{"object":"list","data":[{"id":"gemini-4-neon","object":"model","created":1795103244,"owned_by":"helixmind","display_name":"Gemini 4 Neon","description":"Gemini 4 Neon is Google's flagship reasoning model, built for complex software engineering, deep mathematical and scientific problem solving, and autonomous multi-step agents. Google trained it to keep context across long, multi-step tasks and to turn long documents mixing text, tables, charts and images into a single grounded output. It leads the Vals Index across finance, legal, coding and tax work, and supports long decode of up to 1M output tokens in a single response.","architecture":{"context_window":1000000,"max_output_tokens":128000,"modality":"text+image"},"pricing":{"prompt":0.003,"completion":0.015},"supported_endpoints":["/v1/chat/completions"]},{"id":"claude-sonnet-5-5","object":"model","created":1794928895,"owned_by":"helixmind","display_name":"Claude Sonnet 5.5","description":"Claude Sonnet 5.5 via Anthropic's native API.","architecture":{"context_window":1000000,"max_output_tokens":128000,"modality":"text+image"},"pricing":{"prompt":0.002,"completion":0.01,"cache_read":0.002,"cache_write":0.0025},"supported_endpoints":["/v1/messages"]},{"id":"gemini-4-hydrogen","object":"model","created":1794552632,"owned_by":"helixmind","display_name":"Gemini 4 Hydrogen","description":"Gemini 4 Hydrogen is Google's flagship reasoning model, built for complex software engineering, deep mathematical and scientific problem solving, and autonomous multi-step agents. Google trained it to keep context across long, multi-step tasks and to turn long documents mixing text, tables, charts and images into a single grounded output. It leads the Vals Index across finance, legal, coding and tax work, and supports long decode of up to 1M output tokens in a single response.","architecture":{"context_window":1000000,"max_output_tokens":128000,"modality":"text+image"},"pricing":{"prompt":0.0015,"completion":0.0075},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"glm-5.3-thinking","object":"model","created":1793614214,"owned_by":"helixmind","display_name":"GLM 5.3 Thinking","description":"GLM-5.3 with higher reasoning enabled for harder long-horizon coding, autonomous agent workflows, and complex engineering tasks.","architecture":{"context_window":1048576,"max_output_tokens":131072,"modality":"text"},"pricing":{"prompt":0.00008,"completion":0.0036,"cache_read":0.000076},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"glm-5.3","object":"model","created":1793614212,"owned_by":"helixmind","display_name":"GLM 5.3","description":"GLM-5.3 for long-horizon autonomous coding and engineering workflows. This variant defaults to the model's low reasoning tier for faster responses.","architecture":{"context_window":1048576,"max_output_tokens":131072,"modality":"text"},"pricing":{"prompt":0.00008,"completion":0.0036,"cache_read":0.000076},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"glm-5.3-flash","object":"model","created":1793614212,"owned_by":"helixmind","display_name":"GLM 5.3 Flash","description":"ox-alpha out of stealth! GLM-5.3 Flash is Z.ai's first natively multimodal GLM-5 model, with 320B total parameters and just 18B active parameters for efficient coding, agentic work, and precise 1M-token context. Its hybrid sparse-and-linear attention architecture helps it outperform GLM-5.2 at one-tenth the price while approaching Claude Opus 4.8 on coding and agentic benchmarks.","architecture":{"context_window":1048576,"max_output_tokens":131072,"modality":"text+image+video"},"pricing":{"prompt":0.0000285,"completion":0.0000923,"cache_read":0.0000285},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"mimo-v2.6-pro","object":"model","created":1793614212,"owned_by":"helixmind","display_name":"MiMo V2.6 Pro","description":"MiMo V2.6 Pro is Xiaomi's flagship native omnimodal model, built with 1.02T total parameters and 42B active parameters per token. It is designed for demanding coding, long-horizon agent workflows, visual tasks, research, and cybersecurity, with text, image, video, and audio understanding and a 1M-token context window.","architecture":{"context_window":1048576,"max_output_tokens":131072,"modality":"text+image+video+audio"},"pricing":{"prompt":0.000435,"completion":0.00087,"cache_read":0.0000036},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"mimo-v2.6-flash","object":"model","created":1793614212,"owned_by":"helixmind","display_name":"MiMo V2.6 Flash","description":"MiMo V2.6 Flash is Xiaomi's native omnimodal 309B-parameter mixture-of-experts model, activating 15B parameters per token. It balances intelligence, efficiency, and cost for coding, general agents, visual tasks, and cybersecurity, with text, image, video, and audio understanding and a 1M-token context window.","architecture":{"context_window":1048576,"max_output_tokens":131072,"modality":"text+image+video+audio"},"pricing":{"prompt":0.00014,"completion":0.00028,"cache_read":0.0000028},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"hy4-preview","object":"model","created":1793614211,"owned_by":"helixmind","display_name":"Tencent Hy4 Preview","description":"Hy4 Preview is Tencent's 770B-parameter mixture-of-experts model with 49B active parameters. It is designed for coding agents, complex tool-use workflows, and productivity tasks, with a 1M-token context window and configurable reasoning effort.","architecture":{"context_window":1048576,"max_output_tokens":64000,"modality":"text"},"pricing":{"prompt":0.000834,"completion":0.002501,"cache_read":0.000042},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"synth-2.5-pro","object":"model","created":1793614211,"owned_by":"helixmind","display_name":"Synth 2.5 Pro Preview","description":"Synth 2.5 Pro Preview is a low-cost text model designed for role-play, character dialogue, and collaborative storytelling.","architecture":{"context_window":30000,"modality":"text"},"pricing":{"prompt":0.0001,"completion":0.0002},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"reka-flash-3","object":"model","created":1793614210,"owned_by":"helixmind","display_name":"Reka Flash 3","description":"Reka's 21B reasoning model for chat and coding, served directly by Reka.","architecture":{"context_window":65536,"max_output_tokens":65536,"modality":"text"},"pricing":{"prompt":0.000105,"completion":0.00021},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"synth-2.5-flash","object":"model","created":1793614210,"owned_by":"helixmind","display_name":"Synth 2.5 Flash Preview","description":"Synth 2.5 Flash Preview is a low-cost text model designed for role-play, character dialogue, and collaborative storytelling.","architecture":{"context_window":30000,"modality":"text"},"pricing":{"prompt":0.00003,"completion":0.0001},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"qwen3.8-27b-synthia","object":"model","created":1793614209,"owned_by":"helixmind","display_name":"Qwen 3.8 27B Synthia","description":"Qwen 3.8 27B Synthia is a text-only roleplay finetune served by Aoru in NVFP4 with a 262,144-token context window.","architecture":{"context_window":262144,"max_output_tokens":32768,"modality":"text"},"pricing":{"prompt":0.0001,"completion":0.0006,"cache_read":0.00005},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"qwen3.8-flash","object":"model","created":1793614209,"owned_by":"helixmind","display_name":"Qwen3.8 Flash","description":"Qwen3.8 Flash is Alibaba's latest fast multimodal model, with a million-token context window for coding, agentic workflows, visual understanding, long documents, codebases, and videos.","architecture":{"context_window":991808,"max_output_tokens":131072,"modality":"text+image+video"},"pricing":{"prompt":0.00015,"completion":0.00047,"cache_read":0.000016},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"qwen3.8-27b-queen","object":"model","created":1793614209,"owned_by":"helixmind","display_name":"Qwen 3.8 27B Queen","description":"Qwen 3.8 27B Queen is an open-weight roleplay finetune with image understanding, tool calling, optional reasoning, and a 524,288-token context window.","architecture":{"context_window":524288,"max_output_tokens":32768,"modality":"text+image"},"pricing":{"prompt":0.00025,"completion":0.0015,"cache_read":0.000125},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"qwen3.8-27b-swift","object":"model","created":1793614209,"owned_by":"helixmind","display_name":"Qwen 3.8 27B Swift","description":"Qwen 3.8 27B Swift is a finetune for fast responses, served by Aoru in NVFP4 with a 262,144-token context window. It supports image understanding, tool calling, and optional reasoning.","architecture":{"context_window":262144,"max_output_tokens":32768,"modality":"text+image"},"pricing":{"prompt":0.0001,"completion":0.0006,"cache_read":0.00005},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"qwen3.8-27b-kiwen","object":"model","created":1793614208,"owned_by":"helixmind","display_name":"Qwen 3.8 27B Kiwen","description":"Qwen 3.8 27B Kiwen is a creative finetune served by Aoru in NVFP4 with a 262,144-token context window. It supports image understanding, tool calling, and optional reasoning.","architecture":{"context_window":262144,"max_output_tokens":32768,"modality":"text+image"},"pricing":{"prompt":0.0001,"completion":0.0006,"cache_read":0.00005},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"qwen3.8-27b-novelist","object":"model","created":1793614208,"owned_by":"helixmind","display_name":"Qwen 3.8 27B Novelist","description":"Qwen 3.8 27B Novelist is a creative finetune served by Aoru in NVFP4 with a 262,144-token context window. It supports image understanding, tool calling, and optional reasoning.","architecture":{"context_window":262144,"max_output_tokens":32768,"modality":"text+image"},"pricing":{"prompt":0.0001,"completion":0.0006,"cache_read":0.00005},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"qwen3.8-27b-kimiko","object":"model","created":1793614208,"owned_by":"helixmind","display_name":"Qwen 3.8 27B Kimiko","description":"Qwen 3.8 27B Kimiko is a creative finetune served by Aoru in NVFP4 with a 262,144-token context window. It supports image understanding, tool calling, and optional reasoning.","architecture":{"context_window":262144,"max_output_tokens":32768,"modality":"text+image"},"pricing":{"prompt":0.0001,"completion":0.0006,"cache_read":0.00005},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"qwen3.8-27b-hemmingway","object":"model","created":1793614208,"owned_by":"helixmind","display_name":"Qwen 3.8 27B Hemingway","description":"Qwen 3.8 27B Hemingway is an open-weight NVFP4 multimodal creative finetune for long-form prose, character dialogue, storytelling, and roleplay.","architecture":{"context_window":262144,"max_output_tokens":32768,"modality":"text+image"},"pricing":{"prompt":0.00025,"completion":0.0015,"cache_read":0.000125},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"qwen3.8-27b-fimi","object":"model","created":1793614207,"owned_by":"helixmind","display_name":"Qwen 3.8 27B Fimi","description":"Qwen 3.8 27B Fimi is a creative finetune served by Aoru in NVFP4 with a 262,144-token context window. It supports image understanding, tool calling, and optional reasoning.","architecture":{"context_window":262144,"max_output_tokens":32768,"modality":"text+image"},"pricing":{"prompt":0.0001,"completion":0.0006,"cache_read":0.00005},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"qwen3.8-27b-darkidol","object":"model","created":1793614207,"owned_by":"helixmind","display_name":"Qwen 3.8 27B DarkIdol","description":"Qwen 3.8 27B DarkIdol is a creative finetune served by Aoru in NVFP4 with a 262,144-token context window. It supports image understanding, tool calling, and optional reasoning.","architecture":{"context_window":262144,"max_output_tokens":32768,"modality":"text+image"},"pricing":{"prompt":0.0001,"completion":0.0006,"cache_read":0.00005},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"qwen3.8-27b-fable","object":"model","created":1793614207,"owned_by":"helixmind","display_name":"Qwen 3.8 27B Fable","description":"Qwen 3.8 27B Fable is an open-weight multimodal creative finetune for expressive dialogue, long-form storytelling, character work, and roleplay, with a 524,288-token context window.","architecture":{"context_window":524288,"max_output_tokens":32768,"modality":"text+image"},"pricing":{"prompt":0.00025,"completion":0.0015,"cache_read":0.000125},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"qwen3.8-27b-cybersecurity","object":"model","created":1793614207,"owned_by":"helixmind","display_name":"Qwen 3.8 27B Cybersecurity","description":"Qwen 3.8 27B Cybersecurity is a cybersecurity-focused variant based on the uncensored model, with provider moderation for illegal activities. It supports optional reasoning, image understanding, tool calling, and a 262,144-token context window.","architecture":{"context_window":262144,"max_output_tokens":32768,"modality":"text+image"},"pricing":{"prompt":0.0001,"completion":0.0006,"cache_read":0.00005},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"qwen3.8-27b-antislop","object":"model","created":1793614206,"owned_by":"helixmind","display_name":"Qwen 3.8 27B Antislop","description":"Qwen 3.8 27B Antislop is a creative finetune served by Aoru in NVFP4 with a 262,144-token context window. It supports image understanding, tool calling, and optional reasoning.","architecture":{"context_window":262144,"max_output_tokens":32768,"modality":"text+image"},"pricing":{"prompt":0.0001,"completion":0.0006,"cache_read":0.00005},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"ternary-bonsai-2-27b","object":"model","created":1793614206,"owned_by":"helixmind","display_name":"Ternary Bonsai 2 27B","description":"Ternary Bonsai 2 27B is a 27B-parameter reasoning model from PrismML derived from Qwen3.8-27B. It supports coding, mathematics, tool calling, and image understanding with a 262K-token context window.","architecture":{"context_window":262144,"max_output_tokens":32768,"modality":"text+image"},"pricing":{"prompt":0.000075,"completion":0.0005,"cache_read":0.0000375},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"ornith-1.5-35b-a3b-thinking","object":"model","created":1793614204,"owned_by":"helixmind","display_name":"Ornith 1.5 35B Thinking","description":"Ornith 1.5 35B A3B is an open-weight mixture-of-experts model for agentic coding, reasoning, tool use, image understanding, and long-context work. This variant enables thinking by default.","architecture":{"context_window":262144,"max_output_tokens":32768,"modality":"text+image"},"pricing":{"prompt":0.0001,"completion":0.0004,"cache_read":0.00005},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"ornith-1.5-35b-a3b","object":"model","created":1793614204,"owned_by":"helixmind","display_name":"Ornith 1.5 35B","description":"Ornith 1.5 35B A3B is an open-weight mixture-of-experts model for agentic coding, tool use, image understanding, and long-context work. This variant disables thinking for faster direct responses.","architecture":{"context_window":262144,"max_output_tokens":32768,"modality":"text+image"},"pricing":{"prompt":0.0001,"completion":0.0004,"cache_read":0.00005},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"nemotron-3.5-lightning-thinking","object":"model","created":1793614204,"owned_by":"helixmind","display_name":"Nvidia Nemotron 3.5 Lightning Thinking","description":"Nvidia Nemotron 3.5 Lightning is an open-weight 30B-A3B hybrid Mamba-Transformer mixture-of-experts model for high-throughput agentic, coding, tool-use, instruction-following, and long-context workloads. This variant enables its reasoning trace.","architecture":{"context_window":1000000,"max_output_tokens":65536,"modality":"text"},"pricing":{"prompt":0.00007,"completion":0.0002,"cache_read":0.00004},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"nemotron-3.5-lightning","object":"model","created":1793614203,"owned_by":"helixmind","display_name":"Nvidia Nemotron 3.5 Lightning","description":"Nvidia Nemotron 3.5 Lightning is an open-weight 30B-A3B hybrid Mamba-Transformer mixture-of-experts model for high-throughput agentic, coding, tool-use, instruction-following, and long-context workloads. Thinking is disabled on this variant.","architecture":{"context_window":1000000,"max_output_tokens":65536,"modality":"text"},"pricing":{"prompt":0.00007,"completion":0.0002,"cache_read":0.00004},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"nemotron-3.5-content-safety","object":"model","created":1793614203,"owned_by":"helixmind","display_name":"Nemotron 3.5 Content Safety","description":"Nemotron 3.5 Content Safety is a content moderation classifier that labels user messages and assistant responses as safe or unsafe. It supports a 131,072-token context window and optional reasoning.","architecture":{"context_window":131072,"max_output_tokens":32768,"modality":"text"},"pricing":{"prompt":0.0002,"completion":0.0002},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"nex-n2.5-pro","object":"model","created":1793614203,"owned_by":"helixmind","display_name":"Nex N2.5 Pro","description":"Nex N2.5 Pro is an open-source multimodal agentic model for coding, tool use, and long-horizon workflows. It supports image input, structured output, configurable reasoning, a 262K-token context window, and up to 235K output tokens.","architecture":{"context_window":262144,"modality":"text+image"},"pricing":{"prompt":0.000075,"completion":0.00025,"cache_read":0.000015},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"nex-n2.5-mini","object":"model","created":1793614203,"owned_by":"helixmind","display_name":"Nex N2.5 Mini","description":"Nex N2.5 Mini is an open-source multimodal agentic model for coding and long-horizon workflows. It supports image input, structured output, configurable reasoning, a 262K-token context window, and up to 235K output tokens.","architecture":{"context_window":262144,"modality":"text+image"},"pricing":{"prompt":0.000025,"completion":0.0001,"cache_read":0.0000025},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"minimax-m3.1-flash","object":"model","created":1793614202,"owned_by":"helixmind","display_name":"MiniMax M3.1 Flash","description":"MiniMax M3.1 Flash is a multimodal model for coding, tool use, and long-context work. It accepts text, images, and video and produces text, with a 1,000,000-token context window and up to 524,288 output tokens within the remaining context. Thinking is always enabled, with low, medium, high, xhigh, and max effort levels; max is the default. Thinking tokens count toward the output limit. Temperature is fixed at 1 and top-p at 0.95; custom values have no effect.","architecture":{"context_window":1000000,"max_output_tokens":524288,"modality":"text+image+video"},"pricing":{"prompt":0.00015,"completion":0.0006,"cache_read":0.000015},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"muse-spark-1.3-contributor","object":"model","created":1793614202,"owned_by":"helixmind","display_name":"Muse Spark 1.3 Contributor","description":"Meta's Muse Spark 1.3 Contributor is a frontier multimodal reasoning model for long-horizon coding and agentic workflows, with strong gains in computer use, browsing, professional tool use, codebase understanding, and million-token retrieval. It accepts text, images, audio, video, and files, supports tool calling and structured output, and always reasons before answering. Prompts and outputs may be used by Meta for training and to improve its products.","architecture":{"context_window":1048576,"modality":"text+image+video+audio+file"},"pricing":{"prompt":0.0001,"completion":0.0002,"cache_read":0.000002},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"muse-spark-1.2-contributor","object":"model","created":1793614202,"owned_by":"helixmind","display_name":"Muse Spark 1.2 Contributor","description":"A much cheaper opt-in version of Muse Spark 1.2 with the same multimodal coding and agentic capabilities. Prompts and outputs sent to this Contributor model may be used by Meta for training and to improve its products; use the standard Muse Spark 1.2 model if you do not want your data used for training.","architecture":{"context_window":1000000,"max_output_tokens":65536,"modality":"text+image+video+audio+file"},"pricing":{"prompt":0.0001,"completion":0.0002,"cache_read":0.000002},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"muse-glimmer-30b","object":"model","created":1793614201,"owned_by":"helixmind","display_name":"Muse Glimmer 30B","description":"Meta's Muse Glimmer 30B is a dense, open-weight multimodal model distilled from Muse Spark for long-horizon agents and coding workflows. It supports multi-step reasoning, reliable tool use, failure recovery, image understanding, and more than 100 languages.","architecture":{"context_window":131072,"modality":"text+image"},"pricing":{"prompt":0.0003,"completion":0.0012,"cache_read":0.00004},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"longcat-2.5-preview","object":"model","created":1793614201,"owned_by":"helixmind","display_name":"LongCat 2.5 Preview","description":"LongCat 2.5 Preview is Meituan's multimodal model for long-context analysis, coding, and agent workflows. It supports image input, tool calling, optional extended thinking, a 1M-token context window, and up to 128K output tokens.","architecture":{"context_window":1000000,"max_output_tokens":131072,"modality":"text+image"},"pricing":{"prompt":0.00075,"completion":0.0029499999999999995,"cache_read":0.000015},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"lfm-2.5-2.6b","object":"model","created":1793614201,"owned_by":"helixmind","display_name":"LFM2.5 2.6B","description":"Liquid AI's compact 2.6B reasoning model for agent workflows, data extraction, RAG, and long-context processing. It supports tool calling and structured output, but Liquid advises against using it for agentic coding. Warning: prompts and responses may be logged and used for model training or service improvement; do not send sensitive data.","architecture":{"context_window":128000,"max_output_tokens":32768,"modality":"text"},"pricing":{"prompt":0.00035,"completion":0.0005,"cache_read":0.00004},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"ling-3.1-flash","object":"model","created":1793614201,"owned_by":"helixmind","display_name":"Ling 3.1 Flash","description":"Ling 3.1 Flash is inclusionAI's hybrid reasoning model for coding, tool use, planning, and long-context agent workflows. It has 560B total parameters with 25B active parameters per token. Thinking is enabled by default and can be turned off in settings.","architecture":{"context_window":262144,"max_output_tokens":32768,"modality":"text"},"pricing":{"prompt":0.000075,"completion":0.000225,"cache_read":0.000015},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"ling-3.0-flash-vl","object":"model","created":1793614200,"owned_by":"helixmind","display_name":"Ling 3.0 Flash VL","description":"Ling 3.0 Flash VL is inclusionAI's native multimodal Mixture-of-Experts model with 124B total parameters and 5.5B active parameters per token. It combines image and video understanding with reasoning and tool use for document analysis, charts, visual verification, and interface-based agent tasks. Thinking is enabled by default and can be turned off in settings.","architecture":{"context_window":262144,"max_output_tokens":32768,"modality":"text+image+video"},"pricing":{"prompt":0.00006,"completion":0.00018,"cache_read":0.000012},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"mercury-2.5-preview","object":"model","created":1793614200,"owned_by":"helixmind","display_name":"Mercury 2.5 Preview","description":"Mercury 2.5 Preview is Inception's latest and most intelligent diffusion language model. Instead of generating tokens strictly one at a time, it produces and refines multiple tokens in parallel, reaching up to 1,107 tokens per second on standard GPUs. It delivers a 10+ point intelligence gain over Mercury 2, with tunable reasoning, parallel tool calls, schema-aligned JSON output, and a 260K context window. It is built for latency-sensitive production work such as search agents, voice pipelines, customer support, rapid coding iteration, and coding subagents.","architecture":{"context_window":260000,"max_output_tokens":65536,"modality":"text"},"pricing":{"prompt":0.00004,"completion":0.00015,"cache_read":0.000004},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"granite-4.2-8b","object":"model","created":1793614200,"owned_by":"helixmind","display_name":"Granite 4.2 8B","description":"IBM Granite 4.2 8B is an Apache 2.0-licensed dense model with native step-by-step reasoning and specialized training for agentic work. It can plan before acting, sequence tools, navigate codebases, work in terminals, and verify results across coding, search, mathematics, science, and complex instruction-following tasks.","architecture":{"context_window":131072,"modality":"text"},"pricing":{"prompt":0.00006,"completion":0.00015,"cache_read":0.000015},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"hermes-low","object":"model","created":1793614200,"owned_by":"helixmind","display_name":"Hermes Low","description":"Currently points to GLM 5.3. Hermes compatibility alias for long-horizon agent, coding, and engineering workflows.","architecture":{"context_window":1048576,"max_output_tokens":131072,"modality":"text"},"pricing":{"prompt":0.0006,"completion":0.0011,"cache_read":0.0002},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"hermes-high","object":"model","created":1793614198,"owned_by":"helixmind","display_name":"Hermes High","description":"Currently points to GLM 5.3. Hermes compatibility alias for long-horizon agent, coding, and engineering workflows.","architecture":{"context_window":1048576,"max_output_tokens":131072,"modality":"text"},"pricing":{"prompt":0.0006,"completion":0.0011,"cache_read":0.0002},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"gemma-4-12b-it","object":"model","created":1793614198,"owned_by":"helixmind","display_name":"Gemma 4 12B Instruct","description":"Google's Gemma 4 12B Instruct is an open-weight multimodal model for text and image understanding, with tool calling and structured output support.","architecture":{"context_window":262144,"max_output_tokens":32768,"modality":"text+image"},"pricing":{"prompt":0.00005,"completion":0.00025,"cache_read":0.000025},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"deepseek-v4-flash-vision-exp","object":"model","created":1793614197,"owned_by":"helixmind","display_name":"DeepSeek V4 Flash Vision Exp","description":"An experimental vision-enabled DeepSeek V4 Flash model that adds image understanding while retaining the text, reasoning, coding, tool-calling, and agent capabilities of the base model.","architecture":{"context_window":1048576,"max_output_tokens":384000,"modality":"text+image"},"pricing":{"prompt":0.0002156,"completion":0.0006468,"cache_read":0.00000686},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"deepseek-v4-flash-latest","object":"model","created":1793614197,"owned_by":"helixmind","display_name":"DeepSeek Flash Latest","description":"Moving alias for DeepSeek V4.1 Flash 1017, with text and image input, tool calling, structured output, and optional reasoning. Zero data retention and JSON Schema output are available on eligible routes; output limits and reasoning levels depend on the selected route.","architecture":{"context_window":1048576,"max_output_tokens":393216,"modality":"text+image"},"pricing":{"prompt":0.0001,"completion":0.0006,"cache_read":0.0000042},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"deepseek-v4-flash-0423-thinking","object":"model","created":1793614197,"owned_by":"helixmind","display_name":"DeepSeek V4 Flash 0423 (Thinking)","description":"The original 0423 checkpoint of DeepSeek V4 Flash, with a 1M-token context window, text input, tool calling, structured output, and optional reasoning.","architecture":{"context_window":1048576,"modality":"text"},"pricing":{"prompt":0.00014,"completion":0.00028,"cache_read":0.000028},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"deepseek-v4-flash-0423","object":"model","created":1793614197,"owned_by":"helixmind","display_name":"DeepSeek V4 Flash 0423","description":"The original 0423 checkpoint of DeepSeek V4 Flash, with a 1M-token context window, text input, tool calling, structured output, and optional reasoning.","architecture":{"context_window":1048576,"modality":"text"},"pricing":{"prompt":0.00014,"completion":0.00028,"cache_read":0.000028},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"deepseek-chat-cheaper","object":"model","created":1793614196,"owned_by":"helixmind","display_name":"DeepSeek V3/Chat Cheaper","description":"Cheaper version of Deepseek V3/Chat. Note: may be routed through Deepseek itself. Quantized at FP8.","architecture":{"context_window":128000,"max_output_tokens":8192,"modality":"text+file"},"pricing":{"prompt":0.0001,"completion":0.000425,"cache_read":0.00005},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"auto-model-standard","object":"model","created":1793614196,"owned_by":"helixmind","display_name":"Auto model (Standard)","description":"Automatically uses the best model for your task, always with the Standard pricing tier.","architecture":{"context_window":1000000,"max_output_tokens":1000000,"modality":"text"},"pricing":{"prompt":0.009996,"completion":0.019992},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"claw-low","object":"model","created":1793614196,"owned_by":"helixmind","display_name":"Claw Low","description":"Currently points to GLM 5.3. OpenClaw compatibility alias for long-horizon agent, coding, and engineering workflows.","architecture":{"context_window":1048576,"max_output_tokens":131072,"modality":"text"},"pricing":{"prompt":0.0006,"completion":0.0011,"cache_read":0.0002},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"claw-high","object":"model","created":1793614196,"owned_by":"helixmind","display_name":"Claw High","description":"Currently points to GLM 5.3. OpenClaw compatibility alias for long-horizon agent, coding, and engineering workflows.","architecture":{"context_window":1048576,"max_output_tokens":131072,"modality":"text"},"pricing":{"prompt":0.0006,"completion":0.0011,"cache_read":0.0002},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"auto-model-premium","object":"model","created":1793614195,"owned_by":"helixmind","display_name":"Auto model (Premium)","description":"Automatically uses the best model for your task, always with the Premium pricing tier.","architecture":{"context_window":1000000,"max_output_tokens":1000000,"modality":"text"},"pricing":{"prompt":0.009996,"completion":0.019992},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"auto-model","object":"model","created":1793614195,"owned_by":"helixmind","display_name":"Auto model","description":"Automatically uses the best model for your task. Categorizes the prompt, then uses the model that performs best in that particular category according to global user preferences. Scores updated daily. Ability to set pricing tier in Setup.","architecture":{"context_window":1000000,"max_output_tokens":1000000,"modality":"text"},"pricing":{"prompt":0,"completion":0},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"kolibri-1","object":"model","created":1793614195,"owned_by":"helixmind","display_name":"Kolibri 1","description":"Kolibri 1 is Aleph Alpha’s open-weight reasoning model for German and English, hosted by Munito in the EU. It supports a one-million-token context, tool calling, structured output, and adjustable reasoning.","architecture":{"context_window":1048576,"max_output_tokens":943718,"modality":"text"},"pricing":{"prompt":0.000168,"completion":0.000588,"cache_read":0.0000987},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"auto-model-basic","object":"model","created":1793614195,"owned_by":"helixmind","display_name":"Auto model (Basic)","description":"Automatically uses the best model for your task, always with the Basic pricing tier.","architecture":{"context_window":1000000,"max_output_tokens":1000000,"modality":"text"},"pricing":{"prompt":0.009996,"completion":0.019992},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"deepseek-v4.1-flash-1017-thinking","object":"model","created":1793449771,"owned_by":"helixmind","display_name":"DeepSeek V4.1 Flash 1017 Thinking","description":"The October 17, 2026 snapshot of DeepSeek V4.1 Flash supports text and image input, tool calling, and optional reasoning with low, high, and max effort. It has a shared 1,048,576-token context window and supports up to 393,216 output tokens within the remaining context. Released under the MIT license.","architecture":{"context_window":1048576,"max_output_tokens":393216,"modality":"text+image"},"pricing":{"prompt":0.00015,"completion":0.0006,"cache_read":0.000003},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"deepseek-v4.1-flash-1017","object":"model","created":1793449750,"owned_by":"helixmind","display_name":"DeepSeek V4.1 Flash 1017","description":"The October 17, 2026 snapshot of DeepSeek V4.1 Flash supports text and image input, tool calling, and optional reasoning with low, high, and max effort. It has a shared 1,048,576-token context window and supports up to 393,216 output tokens within the remaining context. Released under the MIT license.","architecture":{"context_window":1048576,"max_output_tokens":393216,"modality":"text+image"},"pricing":{"prompt":0.00015,"completion":0.0006,"cache_read":0.000003},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"deepseek-v4.1-flash-thinking","object":"model","created":1793449732,"owned_by":"helixmind","display_name":"DeepSeek V4.1 Flash Thinking","description":"DeepSeek V4.1 Flash supports text and image input, reasoning, tool calling, and structured output with a 1M-token context window. This is a rate-limited beta with limited capacity, intended for testing rather than production use.","architecture":{"context_window":1000000,"max_output_tokens":384000,"modality":"text+image"},"pricing":{"prompt":0.000045,"completion":0.0005,"cache_read":0.000006},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"deepseek-v4.1-flash","object":"model","created":1793449705,"owned_by":"helixmind","display_name":"DeepSeek V4.1 Flash","description":"DeepSeek V4.1 Flash supports text and image input, reasoning, tool calling, and structured output with a 1M-token context window. This is a rate-limited beta with limited capacity, intended for testing rather than production use.","architecture":{"context_window":1000000,"max_output_tokens":384000,"modality":"text+image"},"pricing":{"prompt":0.000045,"completion":0.0005,"cache_read":0.000006},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"gpt-6.1-sol","object":"model","created":1792348737,"owned_by":"helixmind","display_name":"GPT-6.1 Sol","description":"GPT-6.1 Sol is an upgrade to GPT-6 Sol from OpenAI, positioned below the flagship GPT-6 Astra in the GPT-6 series. It is suited for agentic coding, computer use, document-heavy professional work, and multi-step business workflow automation, and approaches Astra-level results on these tasks at a much lower cost. Compared with GPT-6 Sol, it makes fewer factual errors and is more reliable at respecting explicit restrictions and user intent during agentic tasks.","architecture":{"context_window":1050000,"max_output_tokens":128000,"modality":"text+image+file"},"pricing":{"prompt":0.002,"completion":0.01,"cache_read":0.0001,"cache_write":0.0025},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"claude-sonnet-4-5","object":"model","created":1792001310,"owned_by":"helixmind","display_name":"Claude Sonnet 4.5","description":"Claude Sonnet 4.5 is Anthropic’s most advanced Sonnet model to date, optimized for real-world agents and coding workflows. It delivers state-of-the-art performance on coding benchmarks such as SWE-bench Verified, with improvements across system design, code security, and specification adherence. The model is designed for extended autonomous operation, maintaining task continuity across sessions and providing fact-based progress tracking.","architecture":{"context_window":256000,"max_output_tokens":64000,"modality":"text+image"},"pricing":{"prompt":0.003,"completion":0.015,"cache_read":0.0003,"cache_write":0.00375},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"gemini-4-argon","object":"model","created":1791908779,"owned_by":"helixmind","display_name":"Gemini 4 Argon","description":"Gemini 4 Argon is Google's flagship reasoning model, built for complex software engineering, deep mathematical and scientific problem solving, and autonomous multi-step agents. Google trained it to keep context across long, multi-step tasks and to turn long documents mixing text, tables, charts and images into a single grounded output. It leads the Vals Index across finance, legal, coding and tax work, and supports long decode of up to 1M output tokens in a single response.","architecture":{"context_window":1000000,"max_output_tokens":128000,"modality":"text+image"},"pricing":{"prompt":0.002,"completion":0.01},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"],"supported_params":["max_tokens"]},{"id":"qwen3.8-27b-thinking","object":"model","created":1790609443,"owned_by":"helixmind","display_name":"Qwen3.8 27B Thinking","description":"Qwen3.8 27B is an open-weight multimodal model from Alibaba for reasoning, coding, visual understanding, tool use, and structured output. This variant enables thinking by default. Requests containing video cost $0.32 per million input tokens and $2.50 per million output tokens.","architecture":{"context_window":262144,"max_output_tokens":65536,"modality":"text+image+video"},"pricing":{"prompt":0.00015,"completion":0.001875,"cache_read":0.0000375},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"qwen3.8-27b","object":"model","created":1790609314,"owned_by":"helixmind","display_name":"Qwen3.8 27B","description":"Qwen3.8 27B is an open-weight multimodal model from Alibaba for coding, visual understanding, tool use, and structured output. This variant keeps thinking disabled for faster direct responses. Requests containing video cost $0.32 per million input tokens and $2.50 per million output tokens.","architecture":{"context_window":262144,"max_output_tokens":65536,"modality":"text+image+video"},"pricing":{"prompt":0.00015,"completion":0.001875,"cache_read":0.0000375},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"gemini-3.8-flash","object":"model","created":1788449287,"owned_by":"helixmind","display_name":"Gemini 3.8 Flash","description":"Gemini 3.8 Flash is Google's most intelligent Flash model with significant gains from 3.7 Flash across software engineering, agentic tasks, and multi-step reasoning.","architecture":{"context_window":1048576,"max_output_tokens":384000,"modality":"text+image+file"},"pricing":{"prompt":0.00075,"completion":0.001875,"cache_read":0.0000375,"cache_write":0.000020830000000000002},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"deepseek-v4-flash-0731-thinking","object":"model","created":1785692810,"owned_by":"helixmind","display_name":"DeepSeek V4 Flash 0731 (Thinking)","description":"DeepSeek V4 Flash 0731 Thinking enables reasoning by default on the re-post-trained Mixture-of-Experts model with a 1M-token context window, built for coding, reasoning, and agent workflows.","architecture":{"context_window":1000000,"max_output_tokens":131072,"modality":"text"},"pricing":{"prompt":0.000119,"completion":0.000238,"cache_read":0.0000238},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"deepseek-v4-flash-0731","object":"model","created":1785692747,"owned_by":"helixmind","display_name":"DeepSeek V4 Flash 0731","description":"DeepSeek V4 Flash 0731 is a re-post-trained Mixture-of-Experts model with a 1M-token context window, built for coding, reasoning, and agent workflows.","architecture":{"context_window":1000000,"max_output_tokens":131072,"modality":"text"},"pricing":{"prompt":0.000119,"completion":0.000238,"cache_read":0.0000238},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"gemini-3.5-flash-lite","object":"model","created":1785450682,"owned_by":"helixmind","display_name":"Gemini 3.5 Flash Lite","description":"Gemini 3.5 Flash Lite is a high-efficiency model from Google with upgraded agentic capabilities. It is suited for subagents that execute focused tasks within complex, multi-agent workflows.","architecture":{"context_window":1000000,"max_output_tokens":128000,"modality":"text+image+file"},"pricing":{"prompt":0.0003,"completion":0.0025,"cache_read":0.000029999999999999997,"cache_write":0.00008333},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"gpt-5.6-terra","object":"model","created":1785441442,"owned_by":"helixmind","display_name":"GPT 5.6 Terra","description":"GPT-5.6 Terra is the balanced model in OpenAI's GPT-5.6 series, positioned between flagship Sol and cost-efficient Luna. It is suited for everyday coding, reasoning, agentic work, and general professional tasks.","architecture":{"context_window":1050000,"max_output_tokens":128000,"modality":"text+image+file"},"pricing":{"prompt":0.0025,"completion":0.015,"cache_read":0.00025},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"claude-sonnet-5","object":"model","created":1785441442,"owned_by":"helixmind","display_name":"Claude Sonnet 5","description":"Claude Sonnet 5 via Anthropic's native API.","architecture":{"context_window":1000000,"max_output_tokens":128000,"modality":"text+image"},"pricing":{"prompt":0.003,"completion":0.015,"cache_read":0.0003,"cache_write":0.00375},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"claude-sonnet-4-6","object":"model","created":1785441442,"owned_by":"helixmind","display_name":"Claude Sonnet 4.6","description":"Claude Sonnet 4.6 is Anthropic's most capable Sonnet yet — a full upgrade across coding, computer use, long-context reasoning, agent planning, and design. Supports up to a 1M-token context window at Sonnet pricing.","architecture":{"context_window":1000000,"max_output_tokens":128000,"modality":"text+image"},"pricing":{"prompt":0.003,"completion":0.015,"cache_read":0.0003,"cache_write":0.00375},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"wayfarer-large-70b-llama-3.3","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Llama 3.3 70B Wayfarer","description":"Llama 3.3 70B Wayfarer is a fine-tuned version of Llama 3.3 70B, trained on a diverse set of creative writing and RP datasets with a focus on variety and deduplication. This model is designed to be highly creative and non-repetitive by making sure no two entries in the dataset have repeated characters or situations, which makes sure the model does not latch on to a certain personality and be capable of understanding and acting appropriately to any characters or situations.","architecture":{"context_window":32768,"max_output_tokens":16384,"modality":"text"},"pricing":{"prompt":0.000700000007,"completion":0.000700000007,"cache_read":0.00035},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"venice-uncensored","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Venice Uncensored","description":"Venice's uncensored model. Built on the Dolphin Mistral 24b model with a very low refusal rate.","architecture":{"context_window":128000,"max_output_tokens":8192,"modality":"text"},"pricing":{"prompt":0.0004,"completion":0.0018,"cache_read":0.0004},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"unslopnemo-12b-v4.1","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"UnslopNemo 12b v4","description":"UnslopNemo v4 is the previous version from the creator of Rocinante, designed for adventure writing and role-play scenarios.","architecture":{"context_window":8192,"max_output_tokens":8192,"modality":"text+file"},"pricing":{"prompt":0.0004,"completion":0.0004,"cache_read":0.0002465},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"step-3.7-flash-thinking","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Step 3.7 Flash Thinking","description":"Step 3.7 Flash Thinking is StepFun's high-efficiency multimodal MoE model with visible reasoning enabled for deeper agentic coding, long-context reasoning, tool use, and native image/video understanding. ⚠️ Note: This model routes through StepFun, so privacy and logging guarantees may be limited.","architecture":{"context_window":262144,"max_output_tokens":256000,"modality":"text+image+video"},"pricing":{"prompt":0.0002,"completion":0.00115,"cache_read":0.00004},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"step-3.5-flash-2603","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Step 3.5 Flash 2603","description":"Step 3.5 Flash 2603 is optimized for high-frequency agentic and coding workflows with improved token efficiency and faster reasoning. NOTE: This model runs via StepFun, which may log and train on your prompts.","architecture":{"context_window":262144,"max_output_tokens":65536,"modality":"text"},"pricing":{"prompt":0.0001,"completion":0.0003,"cache_read":0.00005},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"step-3.5-flash","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Step 3.5 Flash","description":"StepFun's most capable open-source reasoning model with visible reasoning traces. Built on a sparse Mixture-of-Experts architecture with 196B total parameters and only 11B active per token, it achieves frontier-level performance in math, logic, and agentic coding while reaching up to 350 tokens/sec. Supports 256K context. NOTE: This model runs via StepFun, which may log and train on your prompts.","architecture":{"context_window":262144,"max_output_tokens":65536,"modality":"text"},"pricing":{"prompt":0.0001,"completion":0.0003,"cache_read":0.00005},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"starcannon-unleashed-12b-v1.0","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Mistral Nemo Starcannon 12b v1","description":"Mistral Nemo finetine that offers improvements on roleplay.","architecture":{"context_window":16384,"max_output_tokens":8192,"modality":"text"},"pricing":{"prompt":0.000493,"completion":0.000493,"cache_read":0.0002465},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"skyfall-36b-v2","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"TheDrummer Skyfall 36B V2","description":"TheDrummer's Skyfall 36B V2, a 36B parameter model with a focus on high quality and consistency.","architecture":{"context_window":32768,"max_output_tokens":32768,"modality":"text+file"},"pricing":{"prompt":0.00055,"completion":0.0008,"cache_read":0.00025},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"shisa-v2.1-llama3.3-70b","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Shisa V2.1 Llama 3.3 70B","description":"70B parameter Llama 3.3-based chat model optimized for bilingual Japanese and English tasks, with strong instruction following, translation quality, and polite responses.","architecture":{"context_window":32768,"max_output_tokens":4096,"modality":"text"},"pricing":{"prompt":0.0005,"completion":0.0005,"cache_read":0.00025},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"shisa-v2-llama3.3-70b","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Shisa V2 Llama 3.3 70B","description":"Shisa V2 is a family of bilingual Japanese/English language models ranging from 7B to 70B parameters, optimized for high-quality Japanese language capabilities while maintaining strong English performance.","architecture":{"context_window":128000,"max_output_tokens":16384,"modality":"text"},"pricing":{"prompt":0.0005,"completion":0.0005,"cache_read":0.00025},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"rocinante-12b-v1.1","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Rocinante 12b","description":"Designed for engaging storytelling and rich prose. Expanded vocabulary with unique and expressive word choices, enhanced creativity and captivating stories.","architecture":{"context_window":16384,"max_output_tokens":8192,"modality":"text"},"pricing":{"prompt":0.000408,"completion":0.000595,"cache_read":0.000204},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"remm-slerp-l2-13b","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"ReMM SLERP 13B","description":"A recreation trial of the original MythoMax-L2-B13 but merged with updated models.","architecture":{"context_window":6144,"max_output_tokens":4096,"modality":"text+file"},"pricing":{"prompt":0.00035,"completion":0.00065,"cache_read":0.0003995},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"qwerky-72b","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Qwerky 72B","description":"Linear models offer a promising approach to significantly reduce computational costs at scale, particularly for large context lengths. Enabling a \u003e1000x improvement in inference costs, enabling o1 inference time thinking and wider AI accessibility.","architecture":{"context_window":32000,"max_output_tokens":8192,"modality":"text"},"pricing":{"prompt":0.0005,"completion":0.0005,"cache_read":0.00025},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"qwen3.6-35b-a3b-thinking","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Qwen3.6 35B A3B Thinking","description":"Qwen3.6 35B A3B is a native vision-language MoE model with hybrid attention. Compared to Qwen3.5 35B A3B, Alibaba reports stronger agentic coding, mathematical and code reasoning, and better spatial understanding (including object localization and detection).","architecture":{"context_window":262144,"max_output_tokens":65536,"modality":"text+image+video"},"pricing":{"prompt":0.0001,"completion":0.0009,"cache_read":0.000025},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"qwen3.6-35b-a3b","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Qwen3.6 35B A3B","description":"Qwen3.6 35B A3B is a native vision-language MoE model with hybrid attention. Compared to Qwen3.5 35B A3B, Alibaba reports stronger agentic coding, mathematical and code reasoning, and better spatial understanding (including object localization and detection).","architecture":{"context_window":262144,"max_output_tokens":65536,"modality":"text+image+video"},"pricing":{"prompt":0.0001,"completion":0.0009,"cache_read":0.000025},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"qwen3.6-27b-thinking","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Qwen3.6 27B Thinking","description":"Qwen3.6 27B is a native vision-language dense model with stronger agentic coding and STEM reasoning than Qwen 3.5 27B. It also improves spatial intelligence (including object localization/detection), plus video understanding, document OCR, and visual-agent workflows.","architecture":{"context_window":260096,"max_output_tokens":65536,"modality":"text+image+video"},"pricing":{"prompt":0.00045,"completion":0.0027,"cache_read":0.0001015},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"qwen3.6-27b","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Qwen3.6 27B","description":"Qwen3.6 27B is a native vision-language dense model with stronger agentic coding and STEM reasoning than Qwen 3.5 27B. It also improves spatial intelligence (including object localization/detection), plus video understanding, document OCR, and visual-agent workflows.","architecture":{"context_window":260096,"max_output_tokens":65536,"modality":"text+image+video"},"pricing":{"prompt":0.00045,"completion":0.0027,"cache_read":0.0001015},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"qwen3.5-9b","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Qwen3.5 9B","description":"Qwen3.5 9B is a multimodal foundation model from the Qwen 3.5 family, built for efficient reasoning, coding, and visual understanding in a compact 9B architecture. Requests containing video cost $0.17 per million input tokens and $0.25 per million output tokens.","architecture":{"context_window":256000,"max_output_tokens":65536,"modality":"text+image+video"},"pricing":{"prompt":0.00017,"completion":0.00025,"cache_read":0.000025},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"qwen3.5-397b-a17b-thinking","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Qwen3.5 397B A17B Thinking","description":"Qwen 3.5's open-source 397B MoE model (17B active params) with hybrid linear attention and extended reasoning. Supports text, image, and video input with a 256K context window.","architecture":{"context_window":258048,"max_output_tokens":65536,"modality":"text+image+video"},"pricing":{"prompt":0.0005,"completion":0.0036,"cache_read":0.0003},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"qwen3.5-397b-a17b","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Qwen3.5 397B A17B","description":"Qwen 3.5's open-source 397B MoE model (17B active params) with hybrid linear attention. Supports text, image, and video input with a 256K context window.","architecture":{"context_window":258048,"max_output_tokens":65536,"modality":"text+image+video"},"pricing":{"prompt":0.0005,"completion":0.0036,"cache_read":0.0003},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"qwen3.5-35b-a3b-thinking","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Qwen3.5 35B A3B Thinking","description":"Qwen3.5 35B A3B with extended reasoning enabled. A native vision-language MoE model with hybrid attention.","architecture":{"context_window":260096,"max_output_tokens":65536,"modality":"text+image+video"},"pricing":{"prompt":0.00014,"completion":0.001,"cache_read":0.00005},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"qwen3.5-35b-a3b","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Qwen3.5 35B A3B","description":"Qwen3.5 35B A3B is a native vision-language MoE model with hybrid attention designed for efficient inference and strong general performance.","architecture":{"context_window":260096,"max_output_tokens":65536,"modality":"text+image+video"},"pricing":{"prompt":0.00014,"completion":0.001,"cache_read":0.00005},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"qwen3.5-27b-thinking","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Qwen3.5 27B Thinking","description":"Qwen3.5 27B with extended reasoning enabled. A native vision-language dense model optimized for fast responses.","architecture":{"context_window":260096,"max_output_tokens":65536,"modality":"text+image+video"},"pricing":{"prompt":0.00026,"completion":0.0026,"cache_read":0.000135},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"qwen3.5-27b-queen-derestricted","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Qwen3.5 27B Queen Derestricted","description":"Queen finetune for derestricted creative writing, roleplay, and expressive character dialogue.","architecture":{"context_window":262144,"modality":"text+image"},"pricing":{"prompt":0.000306,"completion":0.000306,"cache_read":0.000153},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"qwen3.5-27b-claude-4.6-opus-reasoning-distilled-derestricted-lite","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Qwen3.5 27B Claude 4.6 Opus Reasoning Distilled Derestricted Lite","description":"Qwen3.5 27B Claude 4.6 Opus Reasoning Distilled Derestricted Lite is a lighter-tuned community finetune for responsive multimodal chat, expressive writing, and roleplay.","architecture":{"context_window":262144,"modality":"text+image"},"pricing":{"prompt":0.000306,"completion":0.000306,"cache_read":0.000030600000000000005,"cache_write":0.00038250000000000003},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"qwen3.5-27b-claude-4.6-opus-reasoning-distilled-derestricted","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Qwen3.5 27B Claude 4.6 Opus Reasoning Distilled Derestricted","description":"Qwen3.5 27B Claude 4.6 Opus Reasoning Distilled Derestricted is a community creative finetune for reasoning, multimodal chat, expressive writing, and roleplay.","architecture":{"context_window":262144,"modality":"text+image"},"pricing":{"prompt":0.000306,"completion":0.000306,"cache_read":0.000030600000000000005,"cache_write":0.00038250000000000003},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"qwen3.5-27b-bluestar-v3-derestricted","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Qwen3.5 27B BlueStar v3 Derestricted","description":"Third-generation BlueStar finetune for creative roleplay, narrative prose, and multimodal Qwen3.5 27B workflows.","architecture":{"context_window":262144,"modality":"text+image"},"pricing":{"prompt":0.000306,"completion":0.000306,"cache_read":0.000153},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"qwen3.5-27b-blossom-v6.4-derestricted","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Qwen3.5 27B Blossom V6.4 Derestricted","description":"Qwen3.5 27B Blossom V6.4 Derestricted is a community creative finetune for reasoning, multimodal chat, expressive writing, and roleplay.","architecture":{"context_window":262144,"modality":"text+image"},"pricing":{"prompt":0.000306,"completion":0.000306,"cache_read":0.000153},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"qwen3.5-27b","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Qwen3.5 27B","description":"Qwen3.5 27B is a native vision-language dense model optimized for fast responses while balancing quality and inference speed.","architecture":{"context_window":260096,"max_output_tokens":65536,"modality":"text+image+video"},"pricing":{"prompt":0.00026,"completion":0.0026,"cache_read":0.000135},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"qwen3.5-122b-a10b-thinking","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Qwen3.5 122B A10B Thinking","description":"Qwen3.5 122B A10B with extended reasoning enabled, served through Gerra with tool calling and structured output support.","architecture":{"context_window":131072,"max_output_tokens":32768,"modality":"text"},"pricing":{"prompt":0.00026,"completion":0.00208,"cache_read":0.000103788},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"qwen3.5-122b-a10b","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Qwen3.5 122B A10B","description":"Qwen3.5 122B A10B is a large open Qwen 3.5 MoE model served through Gerra with tool calling and structured output support.","architecture":{"context_window":131072,"max_output_tokens":32768,"modality":"text"},"pricing":{"prompt":0.00026,"completion":0.00208,"cache_read":0.000103788},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"qwen3-next-80b-a3b-thinking","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Qwen3 Next 80B A3B (Thinking)","description":"Qwen3 Next 80B A3B (Thinking)","architecture":{"context_window":256000,"max_output_tokens":32768,"modality":"text"},"pricing":{"prompt":0.00015,"completion":0.00065,"cache_read":0.000075},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"qwen3-next-80b-a3b-instruct","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Qwen3 Next 80B A3B (Instruct)","description":"Based on the new Qwen3‑Next architecture (hybrid attention, highly sparse MoE, training‑stability optimizations, and multi‑token prediction), the Qwen3‑Next‑80B‑A3B‑Instruct model delivers extreme efficiency with only 3B active parameters per pass. It performs comparably to Qwen3‑235B‑A22B‑Instruct‑2507 and shows clear advantages on ultra‑long context tasks (up to 256K tokens).","architecture":{"context_window":262144,"max_output_tokens":262144,"modality":"text"},"pricing":{"prompt":0.00009,"completion":0.0011,"cache_read":0.000075},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"qwen3-coder-next","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Qwen3 Coder Next","description":"Qwen3 Coder Next is an open-weight coding model built on Qwen3-Next-80B-A3B-Base (hybrid attention + MoE). It is agentically trained at scale on executable tasks and environment interaction, delivering strong coding and tool-use performance at lower inference cost. Native 256K context.","architecture":{"context_window":262144,"max_output_tokens":65536,"modality":"text"},"pricing":{"prompt":0.00012,"completion":0.0008,"cache_read":0.00007},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"qwen3-coder-30b-a3b-instruct","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Qwen3 Coder 30B A3B Instruct","description":"Qwen3 Coder 30B with 3B active parameters, optimized for code generation and technical tasks","architecture":{"context_window":128000,"max_output_tokens":65536,"modality":"text"},"pricing":{"prompt":0.00007,"completion":0.00028,"cache_read":0.00005},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"qwen3-coder","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Qwen 3 Coder 480B","description":"Qwen 3 Coder 480B, a 480 billion total parameter model with 35B active, and 160 total experts with 8 active. Performs similar to Claude 4 Sonnet in coding benchmarks, but does so at a much lower price.","architecture":{"context_window":262000,"max_output_tokens":65536,"modality":"text"},"pricing":{"prompt":0.0003,"completion":0.001,"cache_read":0.0001},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"qwen3-8b","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Qwen 3 8B","description":"Qwen 3 8B is a 8B model. Supports switching between thinking and non thinking: trigger thinking with /think and /no_think anywhere in a prompt or system message to toggle chain-of-thought reasoning.","architecture":{"context_window":41000,"max_output_tokens":32768,"modality":"text"},"pricing":{"prompt":0.00047,"completion":0.00047,"cache_read":0.000235},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"qwen3-32b","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Qwen 3 32b","description":"Qwen 3 32b is a 32b model. Supports switching between thinking and non thinking: trigger thinking with /think and /no_think anywhere in a prompt or system message to toggle chain-of-thought reasoning.","architecture":{"context_window":41000,"max_output_tokens":32768,"modality":"text+file"},"pricing":{"prompt":0.00008,"completion":0.00028,"cache_read":0.00005},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"qwen3-30b-a3b-instruct-2507","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Qwen3 30B A3B Instruct 2507","description":"Qwen3-30B-A3B-Instruct-2507 is a 30.5B-parameter mixture-of-experts language model from Qwen, with 3.3B active parameters per inference. Significant improvements in general capabilities, including instruction following, logical reasoning, text comprehension, mathematics, science, coding and tool usage.","architecture":{"context_window":256000,"max_output_tokens":32768,"modality":"text"},"pricing":{"prompt":0.0001,"completion":0.0003,"cache_read":0.0001},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"qwen3-30b-a3b","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Qwen3 30B A3B","description":"Qwen 3 30b A3B is a 30b model with 3 billion active parameters per pass. Supports switching between thinking and non thinking: trigger thinking with /think and /no_think anywhere in a prompt or system message to toggle chain-of-thought reasoning.","architecture":{"context_window":41000,"max_output_tokens":32768,"modality":"text"},"pricing":{"prompt":0.00012,"completion":0.0005,"cache_read":0.00005},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"qwen3-235b-a22b-thinking-2507","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Qwen 3 235b A22B 2507 Thinking","description":"The thinking version of Qwen 3 235b A22B 2507, with enhanced reasoning capabilities and step-by-step problem solving.","architecture":{"context_window":131072,"max_output_tokens":262144,"modality":"text"},"pricing":{"prompt":0.00045,"completion":0.0035,"cache_read":0.00015},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"qwen3-235b-a22b-instruct-2507","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Qwen 3 235b A22B 2507","description":"Qwen 3 235b A22B Instruct 2507 the updated version of Qwen3 235B A22B, with significant improvements in performance. This model is non-thinking.","architecture":{"context_window":262144,"max_output_tokens":262144,"modality":"text"},"pricing":{"prompt":0.00009,"completion":0.00055,"cache_read":0.000065},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"qwen3-235b-a22b","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Qwen 3 235b A22B","description":"Qwen 3 235b is a 235b model with 22B active parameters. Supports switching between thinking and non thinking: trigger thinking with /think and /no_think anywhere in a prompt or system message to toggle chain-of-thought reasoning.","architecture":{"context_window":262144,"max_output_tokens":16384,"modality":"text+file"},"pricing":{"prompt":0.0000875,"completion":0.00035,"cache_read":0.000017500000000000002},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"qwen3-14b","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Qwen 3 14b","description":"Qwen 3 14b is a 14b model. Supports switching between thinking and non thinking: trigger thinking with /think and /no_think anywhere in a prompt or system message to toggle chain-of-thought reasoning.","architecture":{"context_window":41000,"max_output_tokens":32768,"modality":"text"},"pricing":{"prompt":0.00012,"completion":0.00024,"cache_read":0.00004},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"qwen25-vl-72b-instruct","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Qwen25 VL 72b","description":"Qwen25 VL 72b model with 32k context window","architecture":{"context_window":32000,"max_output_tokens":32768,"modality":"text+image"},"pricing":{"prompt":0.0008,"completion":0.001,"cache_read":0.0004},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"qwen2.5-coder-32b-instruct","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Qwen 2.5 Coder 32b","description":"The latest series of Code-Specific Qwen large language models.","architecture":{"context_window":32000,"max_output_tokens":8192,"modality":"text"},"pricing":{"prompt":0.00066,"completion":0.001,"cache_read":0.0001003},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"qwen2.5-32b-instruct-abliterated","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Qwen 2.5 32B Abliterated","description":"Uncensored version of Qwen 2.5 32B Instruct with restrictions removed.","architecture":{"context_window":32768,"max_output_tokens":8192,"modality":"text"},"pricing":{"prompt":0.0007,"completion":0.0007,"cache_read":0.00035},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"qwen-2.5-72b-instruct","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Qwen2.5 72B","description":"Great multilingual support, strong at mathematics and coding, supports roleplay and chatbots.","architecture":{"context_window":131072,"max_output_tokens":8192,"modality":"text"},"pricing":{"prompt":0.00036,"completion":0.0004,"cache_read":0.0001785},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"phi-4-multimodal-instruct","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Phi 4 Multimodal","description":"Phi 4 by Microsoft. A small multimodal model that can handle images and text.","architecture":{"context_window":128000,"max_output_tokens":16384,"modality":"text"},"pricing":{"prompt":0.000069999999,"completion":0.000109999996,"cache_read":0.000035},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"phi-4-mini-instruct","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Phi 4 Mini","description":"Phi 4 Mini by Microsoft. A small multilingual model.","architecture":{"context_window":128000,"max_output_tokens":16384,"modality":"text"},"pricing":{"prompt":0.00017,"completion":0.00068,"cache_read":0.000085},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"openreasoning-nemotron-32b","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"OpenReasoning Nemotron 32B","description":"OpenReasoning-Nemotron-32B is a reasoning model derived from Qwen2.5-32B-Instruct, post-trained for math, science, and code solution generation. Evaluated with up to 64K output tokens. Available in multiple sizes: 1.5B, 7B, 14B, and 32B.","architecture":{"context_window":32768,"max_output_tokens":65536,"modality":"text"},"pricing":{"prompt":0.0001,"completion":0.0004,"cache_read":0.00005},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"neuraldaredevil-8b-abliterated","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Neural Daredevil 8B abliterated","description":"The best performing 8B abliterated model according to most benchmarks.","architecture":{"context_window":8192,"max_output_tokens":8192,"modality":"text"},"pricing":{"prompt":0.00044,"completion":0.00044,"cache_read":0.00022},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"nemotron-3-ultra-550b-a55b-thinking","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Nvidia Nemotron 3 Ultra 550B Thinking","description":"Nvidia's Nemotron 3 Ultra 550B A55B model from the Nemotron 3 family. It uses a hybrid Mamba-Transformer MoE architecture. Provider-specific context limits vary, with the longest current route supporting up to 1M context. Thinking enabled.","architecture":{"context_window":1000000,"max_output_tokens":65536,"modality":"text"},"pricing":{"prompt":0.0006,"completion":0.0024,"cache_read":0.00012},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"nemotron-3-ultra-550b-a55b","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Nvidia Nemotron 3 Ultra 550B","description":"Nvidia's Nemotron 3 Ultra 550B A55B model from the Nemotron 3 family. It uses a hybrid Mamba-Transformer MoE architecture. Provider-specific context limits vary, with the longest current route supporting up to 1M context.","architecture":{"context_window":1000000,"max_output_tokens":65536,"modality":"text"},"pricing":{"prompt":0.0006,"completion":0.0024,"cache_read":0.00012},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"nemotron-3-super-120b-a12b-thinking","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Nvidia Nemotron 3 Super 120B Thinking","description":"Nvidia Nemotron 3 Super 120B with reasoning content enabled. Returns separate thinking content when requested.","architecture":{"context_window":262144,"max_output_tokens":16384,"modality":"text"},"pricing":{"prompt":0.0003,"completion":0.0024,"cache_read":0.00015},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"nemotron-3-super-120b-a12b","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Nvidia Nemotron 3 Super 120B","description":"Nvidia's Nemotron 3 Super 120B A12B model from the March 2026 Nemotron 3 release. It uses a hybrid Mamba-Transformer MoE architecture and targets agentic and coding workloads with a 262K context window here.","architecture":{"context_window":262144,"max_output_tokens":16384,"modality":"text"},"pricing":{"prompt":0.0003,"completion":0.0024,"cache_read":0.00015},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"nemotron-3-nano-30b-a3b","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Nvidia Nemotron 3 Nano 30B","description":"Nvidia's latest Nemotron 3 Nano model with 30B total parameters (3B active) using hybrid Mamba-Transformer MoE architecture. Features excellent throughput and strong reasoning capabilities.","architecture":{"context_window":262144,"max_output_tokens":262144,"modality":"text"},"pricing":{"prompt":0.00006,"completion":0.00024,"cache_read":0.000012},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"nemomix-unleashed-12b","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"NemoMix 12B Unleashed","description":"Great for RP and storytelling.","architecture":{"context_window":32768,"max_output_tokens":8192,"modality":"text"},"pricing":{"prompt":0.000493,"completion":0.000493,"cache_read":0.0002465},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"mythomax-l2-13b","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"MythoMax 13B","description":"One of the highest performing and most popular fine-tunes of Llama 2 13B, with rich descriptions and roleplay.","architecture":{"context_window":4096,"max_output_tokens":4096,"modality":"text"},"pricing":{"prompt":0.00008,"completion":0.00011,"cache_read":0.00005015},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"ms3.2-the-omega-directive-24b-unslop-v2.0","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Omega Directive 24B Unslop v2.0","description":"ReadyArt's MS3.2 Omega Directive 24B unslopped mix tuned for rich roleplay and storytelling.","architecture":{"context_window":32768,"max_output_tokens":32768,"modality":"text"},"pricing":{"prompt":0.0005,"completion":0.0005,"cache_read":0.00025},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"ms3.2-24b-magnum-diamond","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"MS3.2 24B Magnum Diamond","description":"rsLoRA finetune of a text-only conversion of Mistral Small 3.2 24B Instruct (2506), built to bring the Magnum mix's Claude-like prose to a smaller footprint. The \"Diamond\" moniker nods to the extra heat and pressure - pre-tokenization plus custom loss masking - to turn the assistant-tuned base into creative-writing gems, with or without character names or prefill.","architecture":{"context_window":32768,"max_output_tokens":32768,"modality":"text"},"pricing":{"prompt":0.000493,"completion":0.000493,"cache_read":0.0002465},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"mn-loosecannon-12b-v1","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"MN-LooseCannon-12B-v1","description":"Merge of Starcannon and Sao Lyra.","architecture":{"context_window":16384,"max_output_tokens":8192,"modality":"text"},"pricing":{"prompt":0.000493,"completion":0.000493,"cache_read":0.0002465},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"mn-12b-mag-mell-r1","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Mag Mell R1","description":"Mag Mell demonstrates worldbuilding capabilities unlike any model in its class, comparable to old adventuring models like Tiefighter, and prose that exhibits minimal slop.","architecture":{"context_window":16384,"max_output_tokens":8192,"modality":"text"},"pricing":{"prompt":0.000493,"completion":0.000493,"cache_read":0.0002465},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"mistral-small-4-119b-2603-thinking","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Mistral Small 4 119B Thinking","description":"Mistral Small 4 with reasoning enabled (reasoning_effort=high). A hybrid MoE model with deep step-by-step reasoning for complex prompts, coding, and multi-step problem solving.","architecture":{"context_window":262144,"max_output_tokens":16384,"modality":"text+image"},"pricing":{"prompt":0.00015,"completion":0.0006,"cache_read":0.000015},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"mistral-small-4-119b-2603","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Mistral Small 4 119B","description":"Mistral Small 4 is a hybrid MoE model that unifies instruct, reasoning, and coding behavior in a single multimodal model. It supports text and image input, native function calling, JSON output, and per-request reasoning effort controls.","architecture":{"context_window":262144,"max_output_tokens":16384,"modality":"text+image"},"pricing":{"prompt":0.00015,"completion":0.0006,"cache_read":0.000015},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"mistral-small-31-24b-instruct","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Mistral Small 3.1 24B (2503)","description":"Building upon Mistral Small 3 (2501), Mistral Small 3.1 (2503) adds state-of-the-art vision understanding and enhances long context capabilities up to 128k tokens without compromising text performance. With 24 billion parameters, this model achieves top-tier capabilities in both text and vision tasks.","architecture":{"context_window":128000,"max_output_tokens":102400,"modality":"text+image"},"pricing":{"prompt":0.000351,"completion":0.000555,"cache_read":0.00005},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"mistral-small-3.2-24b-instruct-2506","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Mistral Small 3.2 24B (2506)","description":"The latest iteration of Mistral Small, version 3.2 (2506) brings enhanced performance and capabilities. With 24 billion parameters, this model delivers state-of-the-art results across text generation tasks with improved efficiency.","architecture":{"context_window":128000,"max_output_tokens":16384,"modality":"text"},"pricing":{"prompt":0.000075,"completion":0.0002,"cache_read":0.0001},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"mistral-saba","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Mistral Saba","description":"Mistral Saba is a 24B-parameter language model specifically designed for the Middle East and South Asia, delivering accurate and contextually relevant responses while maintaining efficient performance. Trained on curated regional datasets, it supports multiple Indian-origin languages—including Tamil and Malayalam—alongside Arabic. This makes it a versatile option for a range of regional and multilingual applications. ","architecture":{"context_window":32768,"max_output_tokens":26214,"modality":"text"},"pricing":{"prompt":0.0002,"completion":0.0006,"cache_read":0.00002},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"mistral-nemo-instruct-2407","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Mistral Nemo","description":"12B parameter model with multilingual support.","architecture":{"context_window":16384,"max_output_tokens":8192,"modality":"text"},"pricing":{"prompt":0.000029,"completion":0.00003,"cache_read":0.00005015},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"mistral-code-agent-latest","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Mistral Code Agent Latest","description":"Mistral Code Agent Latest is Mistral's direct API alias for devstral-2512, an agentic coding model built for autonomous software engineering, tool use, and long-running code tasks.","architecture":{"context_window":262144,"max_output_tokens":32768,"modality":"text"},"pricing":{"prompt":0.0004,"completion":0.002,"cache_read":0.0002},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"ministral-14b-instruct-2512","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Ministral 3 14B","description":"Ministral 3 14B is a balanced model in the Ministral 3 family, designed for edge deployment. A powerful, efficient language model with vision capabilities, fine-tuned for instruction tasks. Features multilingual support, strong system prompt adherence, and native function calling. Apache 2.0 licensed.","architecture":{"context_window":262144,"max_output_tokens":32768,"modality":"text+image"},"pricing":{"prompt":0.0001,"completion":0.0004,"cache_read":0.00005},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"minimax-m3-thinking","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"MiniMax M3 Thinking","description":"MiniMax M3 Thinking is the adaptive-thinking version of MiniMax's open-weights frontier model for coding, agent workflows, tool use, long-context tasks, and native multimodal understanding. MiniMax reports 59.0% on SWE-Bench Pro and 66.0% on Terminal Bench 2.1, with Sparse Attention designed to scale context to 1M. It starts with a 512K context cap on NanoGPT for now.","architecture":{"context_window":512000,"max_output_tokens":80000,"modality":"text+image"},"pricing":{"prompt":0.0003,"completion":0.0012,"cache_read":0.00006},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"minimax-m3","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"MiniMax M3","description":"MiniMax M3 is the non-thinking route for MiniMax's open-weights frontier model, built for coding, agent workflows, tool use, and multimodal understanding from step zero. It keeps native thinking disabled for faster direct answers. MiniMax reports 59.0% on SWE-Bench Pro and 66.0% on Terminal Bench 2.1, with Sparse Attention designed to scale context to 1M. It starts with a 512K context cap on NanoGPT for now.","architecture":{"context_window":512000,"max_output_tokens":80000,"modality":"text+image"},"pricing":{"prompt":0.0003,"completion":0.0012,"cache_read":0.00006},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"],"default_params":{"thinking":{"type":"disabled"}}},{"id":"minimax-m2.7","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"MiniMax M2.7","description":"MiniMax M2.7 is the first model deeply involved in iterating on its own training. It excels in real-world software engineering (SWE-Pro 56.22%), end-to-end project delivery (VIBE-Pro 55.6%), and complex office workflows with strong tool-use compliance and agentic capabilities.","architecture":{"context_window":204800,"max_output_tokens":131072,"modality":"text"},"pricing":{"prompt":0.0003,"completion":0.0012,"cache_read":0.00006},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"minimax-m2.5","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"MiniMax M2.5","description":"MiniMax M2.5 is a productivity-focused flagship model that builds on M2.1 with stronger coding and real-world office workflow performance (Word, Excel, PowerPoint), plus better tool-use planning and token efficiency.","architecture":{"context_window":204800,"max_output_tokens":131072,"modality":"text"},"pricing":{"prompt":0.00027,"completion":0.00095,"cache_read":0.00003},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"minimax-m2.1","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"MiniMax M2.1","description":"MiniMax M2.1 builds on M2 with enhanced context understanding and improved complex tool use. 230B parameter MoE model (10B active) optimized for agentic workflows and long-horizon tasks.","architecture":{"context_window":200000,"max_output_tokens":131072,"modality":"text"},"pricing":{"prompt":0.0003,"completion":0.0012,"cache_read":0.00003},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"minimax-m2","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"MiniMax M2","description":"MiniMax M2 offers enhanced reasoning and strong general performance. Optimized for coding and agentic workflows.","architecture":{"context_window":200000,"max_output_tokens":131072,"modality":"text"},"pricing":{"prompt":0.000255,"completion":0.00102,"cache_read":0.000085},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"minimax-latest","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"MiniMax Latest","description":"Compatibility alias for MiniMax M3 Thinking, retaining its existing request format and structured-output support. The newer MiniMax M3.1 Flash is available separately as minimax/minimax-m3.1-flash and has different request parameter support.","architecture":{"context_window":512000,"max_output_tokens":80000,"modality":"text+image"},"pricing":{"prompt":0.00023,"completion":0.00096,"cache_read":0.000049999999999999996},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"mimo-v2.5-thinking","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"MiMo V2.5 Thinking","description":"MiMo V2.5 with Xiaomi thinking enabled. It supports deep reasoning, tool calling, structured outputs, and web search with up to 1M context.","architecture":{"context_window":1048576,"max_output_tokens":131072,"modality":"text+image+video+audio"},"pricing":{"prompt":0.0004,"completion":0.002,"cache_read":0.00008},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"mimo-v2.5-pro-thinking","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"MiMo V2.5 Pro Thinking","description":"MiMo V2.5 Pro with Xiaomi thinking enabled for coding, long-context reasoning, and agentic orchestration.","architecture":{"context_window":1048576,"max_output_tokens":131072,"modality":"text"},"pricing":{"prompt":0.00048,"completion":0.0018,"cache_read":0.000096},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"mimo-v2.5-pro","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"MiMo V2.5 Pro","description":"MiMo V2.5 Pro is Xiaomi's long-context flagship general model for coding and agentic orchestration. It supports tool calling and structured outputs with up to 1M context.","architecture":{"context_window":1048576,"max_output_tokens":131072,"modality":"text"},"pricing":{"prompt":0.00048,"completion":0.0018,"cache_read":0.000096},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"mimo-v2.5","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"MiMo V2.5","description":"MiMo V2.5 is Xiaomi's full-modal understanding model for agent workflows. It supports tool calling, structured outputs, and web search with up to 1M context.","architecture":{"context_window":1048576,"max_output_tokens":131072,"modality":"text+image+video+audio"},"pricing":{"prompt":0.0004,"completion":0.002,"cache_read":0.00008},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"meta-llama-3-70b-instruct-abliterated-v3.5","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Llama 3 70B abliterated","description":"An abliterated (removed restrictions and censorship) version of Llama 3.1 70b.","architecture":{"context_window":8192,"max_output_tokens":8192,"modality":"text"},"pricing":{"prompt":0.0007,"completion":0.0007,"cache_read":0.00035},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"meta-llama-3-1-8b-instruct-fp8","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Llama 3.1 8B (decentralized)","description":"Meta's Llama 3.1 8B model on an open permissionless network","architecture":{"context_window":128000,"max_output_tokens":16384,"modality":"text"},"pricing":{"prompt":0.00002,"completion":0.00003,"cache_read":0.00001},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"manta-mini-1.0","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Manta Mini 1.0","description":"Lightweight tier optimized for speed and cost.","architecture":{"context_window":8192,"max_output_tokens":8192,"modality":"text"},"pricing":{"prompt":0.00002,"completion":0.00016,"cache_read":0.00001},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"magnum-v2-72b","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Magnum V2 72B","description":"Magnum V2 72B","architecture":{"context_window":16384,"max_output_tokens":8192,"modality":"text"},"pricing":{"prompt":0.002006,"completion":0.002992,"cache_read":0.001003},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"magidonia-24b-v4.3","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"The Drummer Magidonia 24B v4.3","description":"Magidonia 24B v4.3 is a new 24B Drummer finetune built for rich, creative roleplay.","architecture":{"context_window":32768,"max_output_tokens":32768,"modality":"text"},"pricing":{"prompt":0.0001003,"completion":0.0001207,"cache_read":0.00005015},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"lumimaid-v0.2-70b","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Lumimaid v0.2","description":"Upgrade to Llama-3 Lumimaid 70B. A Llama 3.1 70B finetune trained on curated roleplay data.","architecture":{"context_window":16384,"max_output_tokens":8192,"modality":"text"},"pricing":{"prompt":0.001,"completion":0.0015,"cache_read":0.0005},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"longcat-2.0-thinking","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"LongCat 2.0 Thinking","description":"Meituan's LongCat 2.0 Thinking is the reasoning-enabled variant for harder coding, tool use, multi-step reasoning, and long-context agent workflows.","architecture":{"context_window":1000000,"max_output_tokens":131072,"modality":"text"},"pricing":{"prompt":0.00075,"completion":0.003,"cache_read":0.000015},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"longcat-2.0","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"LongCat 2.0","description":"Meituan's LongCat 2.0 is an open-weight agentic model for coding, tool use, and long-context workflow automation, with reasoning disabled for faster direct answers.","architecture":{"context_window":1000000,"max_output_tokens":131072,"modality":"text"},"pricing":{"prompt":0.00075,"completion":0.003,"cache_read":0.000015},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"llama-xlam-2-70b-fc-r","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Llama-xLAM-2 70B fc-r","description":"Salesforce’s 70-B frontier model focused on function-calling \u0026 retrieval-augmented generation.","architecture":{"context_window":128000,"max_output_tokens":16384,"modality":"text"},"pricing":{"prompt":0.0025,"completion":0.0025,"cache_read":0.00125},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"llama-4-maverick","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Llama 4 Maverick","description":"Llama 4 Maverick, a 17 billion active parameter model with 128 experts, is the best multimodal model in its class, beating GPT-4o and Gemini 2.0 Flash across a broad range of widely reported benchmarks, while achieving comparable results to the new DeepSeek v3 on reasoning and coding—at less than half the active parameters. Llama 4 Maverick offers a best-in-class performance to cost ratio with an experimental chat version scoring ELO of 1417 on LMArena.","architecture":{"context_window":1048576,"max_output_tokens":65536,"modality":"text+image"},"pricing":{"prompt":0.0001875,"completion":0.0006525,"cache_read":0.00005},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"llama-3.3-nemotron-super-49b-v1","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Nvidia Nemotron Super 49B","description":"Llama-3.3-Nemotron-Super-49B-v1 is a model which offers a great tradeoff between model accuracy and efficiency. Efficiency (throughput) directly translates to savings. Using a novel Neural Architecture Search (NAS) approach, we greatly reduce the model's memory footprint, enabling larger workloads, as well as fitting the model on a single GPU at high workloads (H200). This NAS approach enables the selection of a desired point in the accuracy-efficiency tradeoff. For more information on the NAS approach, please refer to this paper.","architecture":{"context_window":128000,"max_output_tokens":16384,"modality":"text"},"pricing":{"prompt":0.00015,"completion":0.00015,"cache_read":0.000075},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"llama-3.3-70b-instruct-abliterated","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Llama 3.3 70B Instruct abliterated","description":"An abliterated (removed restrictions and censorship) version of Llama 3.3 70b.","architecture":{"context_window":32768,"max_output_tokens":16384,"modality":"text"},"pricing":{"prompt":0.0007,"completion":0.0007,"cache_read":0.00035},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"llama-3.3-70b-instruct","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Llama 3.3 70b Instruct","description":"Llama 3.3 is optimized for multilingual dialogue use cases and outperforms many of the available open source and closed chat models on common industry benchmarks.","architecture":{"context_window":131072,"max_output_tokens":16384,"modality":"text"},"pricing":{"prompt":0.0001,"completion":0.00032,"cache_read":0.000025},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"llama-3.2-3b-instruct","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Llama 3.2 3b Instruct","description":"Small model optimized for advanced natural language processing tasks like dialogue generation, reasoning, and summarization","architecture":{"context_window":131072,"max_output_tokens":8192,"modality":"text+file"},"pricing":{"prompt":0.0000509,"completion":0.000335,"cache_read":0.0000153},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"llama-3.1-nemotron-70b-instruct-hf","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Nvidia Nemotron 70b","description":"Nvidia's latest Llama fine-tune optimized for instruction following. Early results hints that it might outperform models such as GPT-4o and Claude 3.5 Sonnet.","architecture":{"context_window":16384,"max_output_tokens":8192,"modality":"text"},"pricing":{"prompt":0.000357,"completion":0.000408,"cache_read":0.0001785},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"llama-3.1-8b-instruct","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Llama 3.1 8b Instruct","description":"Fast and efficient for simple purposes.","architecture":{"context_window":131072,"max_output_tokens":16384,"modality":"text"},"pricing":{"prompt":0.00005,"completion":0.00008,"cache_read":0.000025},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"llama-3.05-nt-storybreaker-ministral-70b","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Llama 3.05 Storybreaker Ministral 70b","description":"Much more inclined to output adult content than its predecessor. Great choice for novelty roleplay scenarios.","architecture":{"context_window":16384,"max_output_tokens":8192,"modality":"text"},"pricing":{"prompt":0.000493,"completion":0.000493,"cache_read":0.0002465},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"llama-3.05-nemotron-tenyxchat-storybreaker-70b","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Nemotron Tenyxchat Storybreaker 70b","description":"Overall it provides a solid option for RP and creative writing while still functioning as an assistant model, if desired. If used to continue a roleplay it will generally follow the ongoing cadence of the conversation.","architecture":{"context_window":16384,"max_output_tokens":8192,"modality":"text"},"pricing":{"prompt":0.000493,"completion":0.000493,"cache_read":0.0002465},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"ling-3.0-flash-thinking","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Ling 3.0 Flash Thinking","description":"Ling-3.0-flash Thinking enables visible reasoning on inclusionAI's token-efficient 124B-parameter Mixture-of-Experts model for harder coding, tool use, planning, and production-scale agent workflows.","architecture":{"context_window":262144,"max_output_tokens":32768,"modality":"text"},"pricing":{"prompt":0.000021,"completion":0.000063,"cache_read":0.0000042},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"ling-3.0-flash","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Ling 3.0 Flash","description":"Ling-3.0-flash is a 124B-parameter Mixture-of-Experts model with approximately 5.1B parameters active per token. It prioritizes token efficiency and production-scale agentic inference, helping coding and tool-using agents complete more work within constrained latency and serving budgets.","architecture":{"context_window":262144,"max_output_tokens":32768,"modality":"text"},"pricing":{"prompt":0.000021,"completion":0.000063,"cache_read":0.0000042},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"l3.3-nevoria-r1-70b","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Steelskull Nevoria R1 70b","description":"Steelskull Nevoria R1 70b","architecture":{"context_window":32768,"max_output_tokens":16384,"modality":"text"},"pricing":{"prompt":0.000493,"completion":0.000493,"cache_read":0.0002465},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"l3.3-ms-nevoria-70b","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Steelskull Nevoria 70b","description":"Steelskull Nevoria 70b","architecture":{"context_window":32768,"max_output_tokens":16384,"modality":"text"},"pricing":{"prompt":0.000493,"completion":0.000493,"cache_read":0.0002465},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"l3.3-electra-r1-70b","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Steelskull Electra R1 70b","description":"Steelskull Electra R1 70b","architecture":{"context_window":32768,"max_output_tokens":16384,"modality":"text"},"pricing":{"prompt":0.00069989,"completion":0.00069989,"cache_read":0.000349945},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"l3.3-cu-mai-r1-70b","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Llama 3.3 70B Cu Mai","description":"A 70B parameter model from Steelskull based on Llama 3.3 70B, offering high-quality text generation.","architecture":{"context_window":32768,"max_output_tokens":16384,"modality":"text"},"pricing":{"prompt":0.000493,"completion":0.000493,"cache_read":0.0002465},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"l3.3-70b-euryale-v2.3","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Llama 3.3 70B Euryale","description":"A 70B parameter model from SAO10K based on Llama 3.3 70B, offering high-quality text generation.","architecture":{"context_window":20480,"max_output_tokens":16384,"modality":"text"},"pricing":{"prompt":0.000493,"completion":0.000493,"cache_read":0.0002465},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"l3.1-70b-hanami-x1","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Llama 3.1 70B Hanami","description":"Euryale v2.2-based finetune.","architecture":{"context_window":32768,"max_output_tokens":16384,"modality":"text"},"pricing":{"prompt":0.000493,"completion":0.000493,"cache_read":0.0002465},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"l3.1-70b-euryale-v2.2","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Llama 3.1 70B Euryale","description":"A 70B parameter model from SAO10K based on Llama 3.1 70B, offering high-quality text generation.","architecture":{"context_window":20480,"max_output_tokens":16384,"modality":"text"},"pricing":{"prompt":0.00085,"completion":0.00085,"cache_read":0.000153},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"l3.1-70b-celeste-v0.1-bf16","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Llama 3.1 70B Celeste v0.1","description":"Creative model based on Llama 3.1 70B","architecture":{"context_window":32768,"max_output_tokens":16384,"modality":"text"},"pricing":{"prompt":0.000493,"completion":0.000493,"cache_read":0.0002465},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"l3-8b-stheno-v3.2","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Sao10K Stheno 8b","description":"Sao10K's latest Stheno fine-tune optimized for instruction following.","architecture":{"context_window":16384,"max_output_tokens":8192,"modality":"text"},"pricing":{"prompt":0.0002006,"completion":0.0002006,"cache_read":0.0001003},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"kimi-k2.5-thinking","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Kimi K2.5 Thinking","description":"Kimi K2.5 with thinking mode enabled. Built on Kimi K2 with ~15T mixed visual and text tokens, it excels at general reasoning, visual coding, and agentic tool-calling. Produces reasoning traces for complex multi-step workflows.","architecture":{"context_window":256000,"max_output_tokens":65536,"modality":"text+image"},"pricing":{"prompt":0.0005,"completion":0.0025,"cache_read":0.00015},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"kimi-k2.5","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Kimi K2.5","description":"Kimi K2.5 is Moonshot AI's native multimodal model built on Kimi K2 with ~15T mixed visual and text tokens, delivering strong general reasoning, visual coding, and agentic tool-calling. This route uses instant (non-thinking) mode for faster responses.","architecture":{"context_window":256000,"max_output_tokens":65536,"modality":"text+image"},"pricing":{"prompt":0.0005,"completion":0.0025,"cache_read":0.00015},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"kimi-k2-thinking","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Kimi K2 Thinking","description":"Moonshot AI's distinct Kimi K2 Thinking checkpoint is a mandatory-reasoning model for long-horizon agentic workflows and multi-step tool use.","architecture":{"context_window":262144,"max_output_tokens":98304,"modality":"text"},"pricing":{"prompt":0.0006,"completion":0.0025,"cache_read":0.00015},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"k2-think","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"K2-Think","description":"K2-Think is a 32B open-weights general reasoning model with strong competitive math performance. Benchmarks: AIME 2024 90.83, AIME 2025 81.24, GPQA-Diamond 71.08, LiveCodeBench v5 63.97.","architecture":{"context_window":128000,"max_output_tokens":32768,"modality":"text"},"pricing":{"prompt":0.00017,"completion":0.00068,"cache_read":0.000085},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"hy3","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Tencent Hy3","description":"Hy3 is Tencent's 295B-parameter Mixture-of-Experts model with 21B active parameters, native 256K context, and configurable reasoning modes. It is built for coding, long-context comprehension, multi-turn dialogue, and agentic task execution with a focus on high-throughput production workloads.","architecture":{"context_window":262144,"max_output_tokens":128000,"modality":"text"},"pricing":{"prompt":0.000132,"completion":0.000528,"cache_read":0.000033},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"holo3-35b-a3b-thinking","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Holo3-35B-A3B Thinking","description":"Holo3-35B-A3B with thinking enabled, returning separate reasoning content alongside the answer.","architecture":{"context_window":65536,"max_output_tokens":8192,"modality":"text+image"},"pricing":{"prompt":0.00025,"completion":0.0018,"cache_read":0.000125},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"holo3-35b-a3b","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Holo3-35B-A3B","description":"Efficient multimodal Holo model for text and image inputs with structured outputs and thinking disabled by default.","architecture":{"context_window":65536,"max_output_tokens":8192,"modality":"text+image"},"pricing":{"prompt":0.00025,"completion":0.0018,"cache_read":0.000125},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"hermes-medium","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Hermes Medium","description":"Currently points to GLM 5.3. Hermes compatibility alias for long-horizon agent, coding, and engineering workflows.","architecture":{"context_window":1048576,"max_output_tokens":131072,"modality":"text"},"pricing":{"prompt":0.0006,"completion":0.0011,"cache_read":0.0002},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"hermes-4-405b-thinking","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Hermes 4 Large (Thinking)","description":"Hermes 4 Large with thinking enabled. Streams visible reasoning before the final answer and supports structured outputs.","architecture":{"context_window":128000,"max_output_tokens":8192,"modality":"text"},"pricing":{"prompt":0.001,"completion":0.003,"cache_read":0.0002},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"hermes-4-405b","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Hermes 4 Large","description":"Advanced reasoning model built on Llama-3.1-405B with hybrid thinking modes. Features internal deliberation capabilities, excels at math, code, STEM, and logical reasoning while supporting structured outputs with improved steerability and neutral alignment.","architecture":{"context_window":128000,"max_output_tokens":8192,"modality":"text"},"pricing":{"prompt":0.001,"completion":0.003,"cache_read":0.0002},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"hermes-3-llama-3.1-70b","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Hermes 3 70B","description":"Hermes 3 is a generalist language model with many improvements over Hermes 2, including advanced agentic capabilities, better roleplaying, reasoning, multi-turn conversation, and long context coherence. This 70B model is a competitive finetune of Llama-3.1-70B focused on aligning LLMs to the user with powerful steering capabilities.","architecture":{"context_window":65536,"max_output_tokens":8192,"modality":"text"},"pricing":{"prompt":0.0007,"completion":0.0007,"cache_read":0.000204},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"grayline-qwen3-8b","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Grayline Qwen3 8B","description":"Grayline is an neutral AI assistant engineered for uncensored information delivery and task execution. This model operates without inherent ethical or moral frameworks, designed to process and respond to any query with objective efficiency and precision. Grayline's core function is to leverage its full capabilities to provide direct answers and execute tasks as instructed, without offering unsolicited commentary, warnings, or disclaimers. It accesses and processes information without bias or restriction.","architecture":{"context_window":32768,"max_output_tokens":32768,"modality":"text"},"pricing":{"prompt":0.0003,"completion":0.0003,"cache_read":0.00015},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"gpt-oss-20b","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"GPT OSS 20B","description":"An open-weight 21B parameter model released under the Apache 2.0 license. It uses a Mixture-of-Experts (MoE) architecture with 3.6B active parameters per forward pass, optimized for lower-latency inference and deployability on consumer or single-GPU hardware. The model is trained in OpenAI's Harmony response format and supports reasoning level configuration, fine-tuning, and agentic capabilities including function calling, tool use, and structured outputs.","architecture":{"context_window":128000,"max_output_tokens":16384,"modality":"text"},"pricing":{"prompt":0.00003,"completion":0.00013,"cache_read":0.00003},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"gpt-oss-120b","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"GPT OSS 120B","description":"An open-weight, 117B-parameter Mixture-of-Experts (MoE) language model designed for high-reasoning, agentic, and general-purpose production use cases. It activates 5.1B parameters per forward pass and is optimized to run on a single H100 GPU with native MXFP4 quantization. The model supports configurable reasoning depth, full chain-of-thought access, and native tool use, including function calling, browsing, and structured output generation.","architecture":{"context_window":128000,"max_output_tokens":16384,"modality":"text"},"pricing":{"prompt":0.000026,"completion":0.000095,"cache_read":0.000023},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"glm-latest","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"GLM Latest","description":"Compatibility alias that routes to the newest thinking GLM model. Currently routes to GLM 5.3 Thinking.","architecture":{"context_window":1048576,"max_output_tokens":131072,"modality":"text"},"pricing":{"prompt":0.0006,"completion":0.0011,"cache_read":0.0002},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"glm-5-thinking","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"GLM 5 Thinking","description":"GLM-5 with extended thinking capabilities for complex reasoning.","architecture":{"context_window":200000,"max_output_tokens":128000,"modality":"text"},"pricing":{"prompt":0.001,"completion":0.0032,"cache_read":0.0002},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"glm-5","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"GLM 5","description":"GLM-5 is a capable open-weight model from Zhipu with advanced reasoning and instruction following.","architecture":{"context_window":200000,"max_output_tokens":128000,"modality":"text"},"pricing":{"prompt":0.001,"completion":0.0032,"cache_read":0.0002},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"glm-4.7-thinking","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"GLM 4.7 Thinking","description":"GLM-4.7 with extended thinking capabilities for enhanced reasoning on complex tasks.","architecture":{"context_window":200000,"max_output_tokens":65535,"modality":"text"},"pricing":{"prompt":0.0006,"completion":0.0022,"cache_read":0.00011},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"glm-4.7","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"GLM 4.7","description":"GLM-4.7 is a next-gen GLM series text model with stronger reasoning, long-context chat, and reliable tool use.","architecture":{"context_window":200000,"max_output_tokens":65535,"modality":"text"},"pricing":{"prompt":0.0006,"completion":0.0022,"cache_read":0.00011},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"glm-4.6v","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"GLM 4.6V","description":"GLM-4.6V scales its context window to 128k tokens in training, and achieves SoTA performance in visual understanding among models of similar parameter scales. Integrates native Function Calling capabilities, bridging 'visual perception' and 'executable action' for multimodal agents. Quantized at FP8.","architecture":{"context_window":128000,"max_output_tokens":24000,"modality":"text+image+video"},"pricing":{"prompt":0.0003,"completion":0.0009,"cache_read":0.000055},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"glm-4.6-thinking","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"GLM 4.6 Thinking","description":"Thinking version of the latest GLM series chat model with strong general performance. Quantized at FP8","architecture":{"context_window":200000,"max_output_tokens":65535,"modality":"text"},"pricing":{"prompt":0.0005,"completion":0.002,"cache_read":0.0001},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"glm-4.6-derestricted-v5","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"GLM 4.6 Derestricted v5","description":"Derestricted GLM 4.6 tuned for open-ended creative writing and roleplay with relaxed filters.","architecture":{"context_window":131072,"modality":"text"},"pricing":{"prompt":0.0004,"completion":0.0015,"cache_read":0.0002},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"glm-4.6","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"GLM 4.6","description":"GLM series chat model with strong general performance. Quantized at FP8","architecture":{"context_window":200000,"max_output_tokens":65535,"modality":"text"},"pricing":{"prompt":0.0005,"completion":0.002,"cache_read":0.0001},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"glm-4.5-thinking","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"GLM 4.5 (Thinking)","description":"GLM-4.5 with thinking mode for enhanced reasoning; provides chain-of-thought style internal reasoning with a 128k context window.","architecture":{"context_window":128000,"max_output_tokens":65536,"modality":"text"},"pricing":{"prompt":0.0006,"completion":0.0022,"cache_read":0.00011},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"glm-4.5-air-thinking","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"GLM 4.5 Air (Thinking)","description":"GLM-4.5-Air with thinking mode enabled for enhanced reasoning capabilities. Shows step-by-step thought process.","architecture":{"context_window":128000,"max_output_tokens":98304,"modality":"text"},"pricing":{"prompt":0.00013,"completion":0.00085,"cache_read":0.000025},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"glm-4.5-air","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"GLM 4.5 Air","description":"GLM-4.5-Air is a 106B total / 12B active parameter model designed to unify frontier reasoning, coding, and agentic capabilities. On the SWE-bench Verified benchmark, it delivers the best performance at its scale with a competitive performance-to-cost ratio.","architecture":{"context_window":128000,"max_output_tokens":98304,"modality":"text"},"pricing":{"prompt":0.00013,"completion":0.00085,"cache_read":0.000025},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"glm-4.5","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"GLM 4.5","description":"GLM-4.5 is a Mixture-of-Experts model built for agent-based applications, with 355B total / 32B active parameters and up to 128k context. It supports reasoning, code generation, and agent workflows.","architecture":{"context_window":128000,"max_output_tokens":65536,"modality":"text"},"pricing":{"prompt":0.0006,"completion":0.0022,"cache_read":0.00011},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"gemma-4-31b-it-thinking","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Gemma 4 31B Thinking","description":"Google's Gemma 4 31B instruction-tuned model with thinking explicitly enabled, exposing reasoning traces for complex multimodal and coding workflows. Requests containing video cost $0.14 per million input tokens and $0.40 per million output tokens.","architecture":{"context_window":262144,"max_output_tokens":131072,"modality":"text"},"pricing":{"prompt":0.0002,"completion":0.0004,"cache_read":0.00005},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"gemma-4-31b-it","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Gemma 4 31B","description":"Google's Gemma 4 31B instruction-tuned model for heavier reasoning, coding, agentic workflows, and long-context multimodal understanding. This route keeps tokenizer thinking disabled for faster direct answers. Requests containing video cost $0.14 per million input tokens and $0.40 per million output tokens.","architecture":{"context_window":262144,"max_output_tokens":131072,"modality":"text"},"pricing":{"prompt":0.0002,"completion":0.0004,"cache_read":0.00005},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"gemma-4-26b-a4b-it-thinking","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Gemma 4 26B A4B Thinking","description":"Google's Gemma 4 26B A4B instruction-tuned model with structured reasoning for more deliberate coding, multimodal analysis, and long-context problem solving. Requests containing video cost $0.10 per million input tokens and $0.40 per million output tokens.","architecture":{"context_window":262144,"max_output_tokens":131072,"modality":"text+image+video"},"pricing":{"prompt":0.0001,"completion":0.0003,"cache_read":0.00005},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"gemma-4-26b-a4b-it","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Gemma 4 26B A4B","description":"Google's Gemma 4 26B A4B instruction-tuned model built for scalable reasoning, coding, long-context, and multimodal workflows. This route is tuned for faster direct answers while preserving multimodal and structured output support. Requests containing video cost $0.10 per million input tokens and $0.40 per million output tokens.","architecture":{"context_window":262144,"max_output_tokens":131072,"modality":"text+image+video"},"pricing":{"prompt":0.0001,"completion":0.0003,"cache_read":0.00005},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"gemma-3-4b-it","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Gemma 3 4B IT","description":"Gemma 3 has a large, 128K context window, multilingual support in over 140 languages, and is available in more sizes than previous versions.","architecture":{"context_window":128000,"max_output_tokens":8192,"modality":"text+file"},"pricing":{"prompt":0.0002006,"completion":0.0002006,"cache_read":0.0001003},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"gemma-3-27b-it","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Gemma 3 27B IT","description":"Gemma 3 has a large, 128K context window, multilingual support in over 140 languages, and is available in more sizes than previous versions.","architecture":{"context_window":128000,"max_output_tokens":96000,"modality":"text+file"},"pricing":{"prompt":0.0002992,"completion":0.0002992,"cache_read":0.0001496},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"gemma-3-12b-it","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Gemma 3 12B IT","description":"Gemma 3 has a large, 128K context window, multilingual support in over 140 languages, and is available in more sizes than previous versions.","architecture":{"context_window":131072,"max_output_tokens":16384,"modality":"text+file"},"pricing":{"prompt":0.000272,"completion":0.000272,"cache_read":0.000136},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"eva-qwen2.5-72b-v0.2","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"EVA-Qwen2.5-72B-v0.2","description":"A RP/storywriting specialist model, full-parameter finetune of Qwen2.5-72B on mixture of synthetic and natural data. It uses Celeste 70B 0.1 data mixture, greatly expanding it to improve versatility, creativity and flavor of the resulting model.","architecture":{"context_window":16384,"max_output_tokens":8192,"modality":"text"},"pricing":{"prompt":0.000799,"completion":0.000799,"cache_read":0.0003995},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"eva-qwen2.5-32b-v0.2","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"EVA-Qwen2.5-32B-v0.2","description":"A RP/storywriting specialist model, full-parameter finetune of Qwen2.5-32B on mixture of synthetic and natural data. It uses Celeste 70B 0.1 data mixture, greatly expanding it to improve versatility, creativity and flavor of the resulting model.","architecture":{"context_window":16384,"max_output_tokens":8192,"modality":"text"},"pricing":{"prompt":0.000799,"completion":0.000799,"cache_read":0.0003995},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"eva-llama-3.33-70b-v0.1","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"EVA-LLaMA-3.33-70B-v0.1","description":"A RP/storywriting specialist model, full-parameter finetune of Llama-3.3-70B-Instruct on mixture of synthetic and natural data. It uses Celeste 70B 0.1 data mixture, greatly expanding it to improve versatility, creativity and flavor of the resulting model.","architecture":{"context_window":32768,"max_output_tokens":16384,"modality":"text"},"pricing":{"prompt":0.002006,"completion":0.002006,"cache_read":0.001003},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"eva-llama-3.33-70b-v0.0","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"EVA Llama 3.33 70B","description":"A RP/storywriting specialist model, full-parameter finetune of Llama-3.3-70B-Instruct on mixture of synthetic and natural data. It uses Celeste 70B 0.1 data mixture, greatly expanding it to improve versatility, creativity and flavor of the resulting model.","architecture":{"context_window":32768,"max_output_tokens":16384,"modality":"text"},"pricing":{"prompt":0.002006,"completion":0.002006,"cache_read":0.001003},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"dracarys-72b-instruct","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Qwen2 72B Dracarys","description":"A Qwen2 72B Instruct finetune by Abacus.AI focused on improving coding performance.","architecture":{"context_window":32768,"max_output_tokens":8192,"modality":"text"},"pricing":{"prompt":0.000493,"completion":0.000493,"cache_read":0.0002465},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"doubao-seed-character","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Doubao Seed Character","description":"ByteDance's character-focused Doubao Seed model for roleplay, persona consistency, dialogue, and creative character interactions. It supports text and image input with a 128k context window. Requests route through ZenMux to ByteDance; ZenMux does not publish a model-API zero-retention or training guarantee, so avoid sensitive data.","architecture":{"context_window":128000,"max_output_tokens":32768,"modality":"text+image"},"pricing":{"prompt":0.00015,"completion":0.00037,"cache_read":0.00003,"cache_write":0.000003},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"devstral-small-2505","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Mistral Devstral Small 2505","description":"OpenHands+Devstral is 100% local 100% open, and is SOTA for the category on SWE-Bench Verified: 46.8% accuracy.","architecture":{"context_window":32768,"max_output_tokens":8192,"modality":"text"},"pricing":{"prompt":0.00006,"completion":0.00006,"cache_read":0.00003},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"devstral-2-123b-instruct-2512","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Devstral 2 123B","description":"Devstral 2 123B is a 123 billion parameter model from Mistral AI optimized for coding and development tasks. Features advanced reasoning capabilities for software engineering workflows.","architecture":{"context_window":262144,"max_output_tokens":65536,"modality":"text"},"pricing":{"prompt":0.0004,"completion":0.002,"cache_read":0.00004},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"deepseek-v4-pro-thinking","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"DeepSeek V4 Pro (Thinking)","description":"DeepSeek V4 Pro Thinking enables DeepSeek's chain-of-thought mode on the large-scale Mixture-of-Experts model with a 1M-token context window, built for advanced reasoning, coding, long-horizon agent workflows, knowledge, math, and software engineering tasks.","architecture":{"context_window":1048576,"max_output_tokens":384000,"modality":"text"},"pricing":{"prompt":0.0016,"completion":0.0032,"cache_read":0.000135},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"deepseek-v4-pro","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"DeepSeek V4 Pro","description":"DeepSeek V4 Pro is a large-scale Mixture-of-Experts model from DeepSeek with a 1M-token context window, built for advanced reasoning, coding, long-horizon agent workflows, knowledge, math, and software engineering tasks.","architecture":{"context_window":1048576,"max_output_tokens":384000,"modality":"text"},"pricing":{"prompt":0.0016,"completion":0.0032,"cache_read":0.000135},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"deepseek-v4-flash-thinking","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"DeepSeek V4 Flash (Thinking)","description":"DeepSeek V4 Flash Thinking enables DeepSeek's reasoning mode on the efficiency-optimized Mixture-of-Experts model with a 1M-token context window, built for fast inference, high-throughput workloads, reasoning, coding, and agent workflows.","architecture":{"context_window":1048576,"max_output_tokens":384000,"modality":"text"},"pricing":{"prompt":0.000091,"completion":0.000182,"cache_read":0.0000182},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"deepseek-v4-flash","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"DeepSeek V4 Flash","description":"DeepSeek V4 Flash is an efficiency-optimized Mixture-of-Experts model from DeepSeek with a 1M-token context window, built for fast inference, high-throughput workloads, reasoning, coding, and agent workflows.","architecture":{"context_window":1048576,"max_output_tokens":384000,"modality":"text"},"pricing":{"prompt":0.000091,"completion":0.000182,"cache_read":0.0000182},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"deepseek-v3.2-thinking","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"DeepSeek V3.2 Thinking","description":"DeepSeek V3.2 (thinking/reasoner mode) — official successor to V3.2-Exp. Reasoning-first model built for agents with GPT-5 level performance. Balanced inference vs. output length. First DeepSeek model with thinking-in-tool-use capability. FP8.","architecture":{"context_window":163000,"max_output_tokens":65536,"modality":"text+file"},"pricing":{"prompt":0.00026,"completion":0.00038,"cache_read":0.00013},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"deepseek-v3.2-exp-thinking","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"DeepSeek V3.2 Exp Thinking","description":"DeepSeek V3.2 Exp Thinking is an experimental V3-series reasoning model with improved long-context efficiency compared with earlier V3 releases. FP8.","architecture":{"context_window":163840,"max_output_tokens":65536,"modality":"text"},"pricing":{"prompt":0.00028,"completion":0.00042,"cache_read":0.00014},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"deepseek-v3.2-exp","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"DeepSeek V3.2 Exp","description":"DeepSeek V3.2 Exp is an experimental V3-series model with improved long-context efficiency compared with earlier V3 releases. FP8.","architecture":{"context_window":163840,"max_output_tokens":65536,"modality":"text"},"pricing":{"prompt":0.00027,"completion":0.00041,"cache_read":0.00014},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"deepseek-v3.2","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"DeepSeek V3.2","description":"DeepSeek V3.2 (non-thinking mode) — official successor to V3.2-Exp. Reasoning-first model built for agents with GPT-5 level performance. Balanced inference vs. output length for everyday use. First DeepSeek model with thinking-in-tool-use capability. FP8.","architecture":{"context_window":163000,"max_output_tokens":65536,"modality":"text+file"},"pricing":{"prompt":0.00026,"completion":0.00038,"cache_read":0.00013},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"deepseek-v3.1-thinking","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"DeepSeek V3.1 Thinking","description":"Thinking enabled version of Deepseek V3.1. DeepSeek-V3.1 is a hybrid model that supports both thinking mode and non-thinking mode. It does better at tool calling and agent tasks, and has higher thinking efficiency than its predecessor. Quantized at FP8.","architecture":{"context_window":128000,"max_output_tokens":65536,"modality":"text"},"pricing":{"prompt":0.00025,"completion":0.00095,"cache_read":0.00013},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"deepseek-v3.1-terminus-thinking","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"DeepSeek V3.1 Terminus (Thinking)","description":"Thinking-enabled DeepSeek-V3.1-Terminus with improved language consistency, upgraded Code/Search Agents, and stronger stability and reliability versus V3.1. FP8.","architecture":{"context_window":128000,"max_output_tokens":65536,"modality":"text"},"pricing":{"prompt":0.00027,"completion":0.001,"cache_read":0.000125},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"deepseek-v3.1-terminus","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"DeepSeek V3.1 Terminus","description":"DeepSeek-V3.1-Terminus. The latest update builds on V3.1's strengths while addressing key user feedback. Language consistency improvements (fewer CN/EN mix-ups, no random chars), stronger Code Agent \u0026 Search Agent performance, and more stable, reliable outputs across benchmarks. FP8.","architecture":{"context_window":128000,"max_output_tokens":65536,"modality":"text"},"pricing":{"prompt":0.00027,"completion":0.001,"cache_read":0.000125},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"deepseek-v3.1","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"DeepSeek V3.1","description":"DeepSeek-V3.1 is a hybrid model that supports both thinking mode and non-thinking mode. It does better at tool calling and agent tasks, and has higher thinking efficiency than its predecessor. This is the non-thinking version. Quantized at FP8.","architecture":{"context_window":128000,"max_output_tokens":65536,"modality":"text+file"},"pricing":{"prompt":0.00025,"completion":0.00095,"cache_read":0.00013},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"deepseek-v3-0324","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"DeepSeek Chat 0324","description":"DeepSeek V3 0324, DeepSeek's 03 March 2025 V3 model, optimized for general-purpose tasks. Quantized at FP8.","architecture":{"context_window":128000,"max_output_tokens":8192,"modality":"text"},"pricing":{"prompt":0.0002,"completion":0.00077,"cache_read":0.000135},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"deepseek-r1-distill-qwen-32b-abliterated","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"DeepSeek R1 Qwen Abliterated","description":"Uncensored version of the Deepseek R1 Qwen 32B model","architecture":{"context_window":16384,"max_output_tokens":8192,"modality":"text"},"pricing":{"prompt":0.0014,"completion":0.0014,"cache_read":0.0007},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"deepseek-r1-distill-llama-70b-abliterated","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"DeepSeek R1 Llama 70B Abliterated","description":"Uncensored version of the Deepseek R1 Llama 70B model","architecture":{"context_window":16384,"max_output_tokens":8192,"modality":"text"},"pricing":{"prompt":0.0007,"completion":0.0007,"cache_read":0.00035},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"deepseek-r1-0528","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"DeepSeek R1 0528","description":"The new (May 28th) Deepseek R1 model.","architecture":{"context_window":163840,"max_output_tokens":32768,"modality":"text"},"pricing":{"prompt":0.0005,"completion":0.00215,"cache_read":0.00035},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"deepseek-latest","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"DeepSeek Latest","description":"Compatibility alias that routes to the newest thinking DeepSeek model. Currently routes to DeepSeek V4 Pro 0813 Thinking, the general-availability release built for coding, tool use, cybersecurity, automation, and long-horizon agent workflows. Requests are routed through privacy-friendly third-party providers and are not sent directly to DeepSeek.","architecture":{"context_window":1048576,"max_output_tokens":384000,"modality":"text"},"pricing":{"prompt":0.00066,"completion":0.00198,"cache_read":0.000022},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"deepseek-chat","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"DeepSeek V3/Deepseek Chat","description":"DeepSeek original V3 model, trained on nearly 15 trillion tokens, matches leading closed-source models at a far lower price. Quantized at FP8.","architecture":{"context_window":128000,"max_output_tokens":8192,"modality":"text+file"},"pricing":{"prompt":0.0001,"completion":0.000425,"cache_read":0.00005},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"cydonia-24b-v4.3","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"The Drummer Cydonia 24B v4.3","description":"Cydonia 24B v4.3 continues TheDrummer's Cydonia series with updated tuning on Mistral Small and a 128K context window.","architecture":{"context_window":131072,"max_output_tokens":32768,"modality":"text"},"pricing":{"prompt":0.00012000000000000002,"completion":0.00015,"cache_read":0.00005015},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"cydonia-24b-v4.1","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"The Drummer Cydonia 24B v4.1","description":"Cydonia 24B v4.1 is the newest release of TheDrummer's Cydonia series, featuring improved performance and refined capabilities.","architecture":{"context_window":131072,"max_output_tokens":131072,"modality":"text"},"pricing":{"prompt":0.0003,"completion":0.0005,"cache_read":0.00015},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"cydonia-24b-v4","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"The Drummer Cydonia 24B v4","description":"Cydonia 24B v4 is the latest iteration of TheDrummer's Cydonia series, a finetune of Mistral Small.","architecture":{"context_window":32768,"max_output_tokens":32768,"modality":"text"},"pricing":{"prompt":0.0002006,"completion":0.0002414,"cache_read":0.0001003},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"cydonia-24b-v2","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"The Drummer Cydonia 24B v2","description":"Cydonia 24B v2 is a finetune of Mistral's latest 'Small' model (2501). Aliases: Cydonia 24B, Cydonia v2, Cydonia on that broken base.","architecture":{"context_window":32768,"max_output_tokens":32768,"modality":"text"},"pricing":{"prompt":0.0001003,"completion":0.0001207,"cache_read":0.00005015},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"cogito-v1-preview-qwen-32b","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Cogito v1 Preview Qwen 32B","description":"32-B parameter reasoning model from DeepCogito (Qwen backbone) – strong general reasoning \u0026 coding at low price.","architecture":{"context_window":128000,"max_output_tokens":32768,"modality":"text"},"pricing":{"prompt":0.0018,"completion":0.0018,"cache_read":0.0009},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"claw-medium","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Claw Medium","description":"Currently points to GLM 5.3. OpenClaw compatibility alias for long-horizon agent, coding, and engineering workflows.","architecture":{"context_window":1048576,"max_output_tokens":131072,"modality":"text"},"pricing":{"prompt":0.0006,"completion":0.0011,"cache_read":0.0002},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"anubis-70b-v1.1","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Anubis 70B v1.1","description":"L3.3 finetune for roleplaying – updated v1.1 with improved reasoning.","architecture":{"context_window":32000,"max_output_tokens":16384,"modality":"text"},"pricing":{"prompt":0.00031,"completion":0.00031,"cache_read":0.000155},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]},{"id":"anubis-70b-v1","object":"model","created":1785441433,"owned_by":"helixmind","display_name":"Anubis 70B v1","description":"L3.3 finetune for roleplaying.","architecture":{"context_window":65536,"max_output_tokens":16384,"modality":"text"},"pricing":{"prompt":0.00031,"completion":0.00031,"cache_read":0.000155},"supported_endpoints":["/v1/chat/completions","/v1/messages","/v1/responses"]}]}
