{"catalog":{"title":"Open-weight AI models: licenses, sizes and hardware","path":"/open-models","updatedAt":"2026-09-29T18:45:59.488Z","fields":[{"id":"developer","label":"Developer","type":"select","options":["Alibaba Qwen","Allen Institute for AI (Ai2)","Cohere","DeepSeek","Google DeepMind","IBM","Meta","Microsoft","MiniMax","Mistral AI","Moonshot AI","NVIDIA","OpenAI","Z.ai (Zhipu AI)"]},{"id":"released","label":"Released","type":"text"},{"id":"params","label":"Total parameters","type":"text"},{"id":"params_b","label":"Total parameters, billions","type":"number","format":{"suffix":"B","maximumFractionDigits":2}},{"id":"active_params","label":"Active parameters per token","type":"text"},{"id":"active_b","label":"Active parameters, billions","type":"number","format":{"suffix":"B","maximumFractionDigits":2}},{"id":"architecture","label":"Architecture","type":"select","options":["Dense","MoE"]},{"id":"context","label":"Context window","type":"text"},{"id":"context_tokens","label":"Context window, tokens","type":"number","format":{"style":"compact"}},{"id":"modalities","label":"Input","type":"text"},{"id":"license","label":"License","type":"text"},{"id":"license_family","label":"License type","type":"select","options":["Apache 2.0","MIT","Modified MIT","Custom license","Non-commercial"]},{"id":"commercial_use","label":"Commercial use","type":"select","options":["Yes","Yes, with conditions","No"]},{"id":"license_terms","label":"License conditions","type":"textarea"},{"id":"precision","label":"Weights precision","type":"text"},{"id":"download_gb","label":"Official weights, GB","type":"number","format":{"suffix":" GB","maximumFractionDigits":1}},{"id":"official_variants","label":"Official quantized or alternative versions","type":"textarea"},{"id":"hf_repo","label":"Hugging Face repository","type":"url"},{"id":"license_url","label":"License text","type":"url"},{"id":"checked","label":"Checked on","type":"date"},{"id":"sources","label":"Sources","type":"textarea"}]},"entries":[{"slug":"deepseek-v4-1-flash","name":"DeepSeek-V4.1-Flash","path":"/open-models/deepseek-v4-1-flash","category":null,"updatedAt":"2026-09-29T18:43:29.949Z","fields":{"developer":"DeepSeek","released":"September 10, 2026","params":"552B + 196B memory","params_b":748,"active_params":"8B–16B","active_b":8,"architecture":"MoE","context":"1M","context_tokens":1000000,"modalities":"Text, image","license":"MIT","license_family":"MIT","commercial_use":"Yes","license_terms":"MIT: keep the copyright and permission notice.","precision":"Mixed FP8 + BF16","download_gb":510.3,"official_variants":null,"hf_repo":"https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash","license_url":"https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash/blob/main/LICENSE","checked":"2026-09-29","sources":"Hugging Face model page · https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash\nLicense text on Hugging Face · https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash/blob/main/LICENSE\nHugging Face weights index · https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash/raw/main/model.safetensors.index.json\nDeepSeek API docs · https://api-docs.deepseek.com/updates"}},{"slug":"glm-5-3-flash","name":"GLM-5.3-Flash","path":"/open-models/glm-5-3-flash","category":null,"updatedAt":"2026-09-29T18:45:29.786Z","fields":{"developer":"Z.ai (Zhipu AI)","released":"August 26, 2026","params":"320B","params_b":320,"active_params":"18B","active_b":18,"architecture":"MoE","context":null,"context_tokens":null,"modalities":"Text, image","license":"MIT","license_family":"MIT","commercial_use":"Yes","license_terms":"MIT: keep the copyright and permission notice.","precision":"FP8","download_gb":328.3,"official_variants":"zai-org/GLM-5.3-Flash-BF16 (full precision)","hf_repo":"https://huggingface.co/zai-org/GLM-5.3-Flash","license_url":"https://huggingface.co/zai-org/GLM-5.3-Flash/blob/main/LICENSE","checked":"2026-09-29","sources":"Hugging Face model card · https://huggingface.co/zai-org/GLM-5.3-Flash/raw/main/README.md\nHugging Face weights index · https://huggingface.co/zai-org/GLM-5.3-Flash/raw/main/model.safetensors.index.json\nGitHub file · https://raw.githubusercontent.com/zai-org/GLM-5/main/README.md\nz.ai announcement · https://z.ai/blog/glm-5.3-flash"}},{"slug":"qwen3-8-flash-next","name":"Qwen3.8-Flash-Next","path":"/open-models/qwen3-8-flash-next","category":null,"updatedAt":"2026-09-29T18:43:06.186Z","fields":{"developer":"Alibaba Qwen","released":"August 26, 2026","params":"125B","params_b":125,"active_params":"6B","active_b":6,"architecture":"MoE","context":"256K","context_tokens":262144,"modalities":"Text, image","license":"Qwen Community License 1.0","license_family":"Custom license","commercial_use":"Yes, with conditions","license_terms":"Products with more than 100 million monthly active users or more than US$20 million in monthly revenue must display the model name. Any model-as-a-service or AI work assistant business needs a separate license from Qwen before commercial use, whatever its revenue. Internal use is exempt.","precision":null,"download_gb":360,"official_variants":"Qwen/Qwen3.8-Flash-Next-FP8","hf_repo":"https://huggingface.co/Qwen/Qwen3.8-Flash-Next","license_url":"https://huggingface.co/Qwen/Qwen3.8-Flash-Next/blob/main/LICENSE","checked":"2026-09-29","sources":"Hugging Face model card · https://huggingface.co/Qwen/Qwen3.8-Flash-Next/raw/main/README.md\nLicense text on Hugging Face · https://huggingface.co/Qwen/Qwen3.8-Flash-Next/raw/main/LICENSE\nHugging Face weights index · https://huggingface.co/Qwen/Qwen3.8-Flash-Next/raw/main/model.safetensors.index.json\nGitHub file · https://raw.githubusercontent.com/QwenLM/Qwen3.8-Flash-Next/main/README.md (news: 2026-08-26)"}},{"slug":"granite-4-2-8b","name":"Granite 4.2 8B","path":"/open-models/granite-4-2-8b","category":null,"updatedAt":"2026-09-29T18:45:06.979Z","fields":{"developer":"IBM","released":"August 25, 2026","params":"8B","params_b":8,"active_params":null,"active_b":null,"architecture":"Dense","context":"128K","context_tokens":128000,"modalities":"Text","license":"Apache 2.0","license_family":"Apache 2.0","commercial_use":"Yes","license_terms":"Apache 2.0, with no user or revenue limits.","precision":"BF16","download_gb":17.6,"official_variants":"ibm-granite/granite-4.2-8b-fp8; ibm-granite/granite-4.2-8b-mxfp4; ibm-granite/granite-4.2-8b-nvfp4; ibm-granite/granite-4.2-8b-GGUF","hf_repo":"https://huggingface.co/ibm-granite/granite-4.2-8b","license_url":"https://www.apache.org/licenses/LICENSE-2.0","checked":"2026-09-29","sources":"Hugging Face model card · https://huggingface.co/ibm-granite/granite-4.2-8b/raw/main/README.md\nHugging Face API · https://huggingface.co/api/models/ibm-granite/granite-4.2-8b?expand=safetensors\nHugging Face API · https://huggingface.co/api/models?author=ibm-granite"}},{"slug":"granite-4-2-30b","name":"Granite 4.2 30B","path":"/open-models/granite-4-2-30b","category":null,"updatedAt":"2026-09-29T18:45:01.620Z","fields":{"developer":"IBM","released":"August 25, 2026","params":"30B","params_b":30,"active_params":null,"active_b":null,"architecture":"Dense","context":"128K","context_tokens":128000,"modalities":"Text","license":"Apache 2.0","license_family":"Apache 2.0","commercial_use":"Yes","license_terms":"Apache 2.0, with no user or revenue limits.","precision":"BF16","download_gb":58.5,"official_variants":"ibm-granite/granite-4.2-30b-fp8; ibm-granite/granite-4.2-30b-mxfp4; ibm-granite/granite-4.2-30b-nvfp4; ibm-granite/granite-4.2-30b-GGUF","hf_repo":"https://huggingface.co/ibm-granite/granite-4.2-30b","license_url":"https://www.apache.org/licenses/LICENSE-2.0","checked":"2026-09-29","sources":"Hugging Face model card · https://huggingface.co/ibm-granite/granite-4.2-30b/raw/main/README.md\nHugging Face weights index · https://huggingface.co/ibm-granite/granite-4.2-30b/raw/main/model.safetensors.index.json\nHugging Face API · https://huggingface.co/api/models?author=ibm-granite"}},{"slug":"glm-5-3","name":"GLM-5.3","path":"/open-models/glm-5-3","category":null,"updatedAt":"2026-09-29T18:45:24.402Z","fields":{"developer":"Z.ai (Zhipu AI)","released":"August 14, 2026","params":"744B","params_b":744,"active_params":"40B","active_b":40,"architecture":"MoE","context":null,"context_tokens":null,"modalities":"Text","license":"GLM-5.3 License","license_family":"Custom license","commercial_use":"Yes, with conditions","license_terms":"Model-as-a-service operators with group revenue above US$10 billion over 12 months must pass a security review by Z.ai before commercial use.","precision":"FP8","download_gb":755.4,"official_variants":"zai-org/GLM-5.3-BF16 (full precision)","hf_repo":"https://huggingface.co/zai-org/GLM-5.3","license_url":"https://huggingface.co/zai-org/GLM-5.3/blob/main/LICENSE","checked":"2026-09-29","sources":"Hugging Face model card · https://huggingface.co/zai-org/GLM-5.3/raw/main/README.md\nLicense text on Hugging Face · https://huggingface.co/zai-org/GLM-5.3/raw/main/LICENSE\nHugging Face API · https://huggingface.co/api/models/zai-org/GLM-5.3?expand=safetensors\nGitHub file · https://raw.githubusercontent.com/zai-org/GLM-5/main/README.md\nz.ai announcement · https://z.ai/blog/glm-5.3"}},{"slug":"qwen3-8-27b","name":"Qwen3.8-27B","path":"/open-models/qwen3-8-27b","category":null,"updatedAt":"2026-09-29T18:42:52.327Z","fields":{"developer":"Alibaba Qwen","released":"August 14, 2026","params":"27B","params_b":27,"active_params":null,"active_b":null,"architecture":"Dense","context":"256K","context_tokens":262144,"modalities":"Text, image, video","license":"Apache 2.0","license_family":"Apache 2.0","commercial_use":"Yes","license_terms":"Apache 2.0, with no user or revenue limits.","precision":null,"download_gb":55.6,"official_variants":"Qwen/Qwen3.8-27B-FP8","hf_repo":"https://huggingface.co/Qwen/Qwen3.8-27B","license_url":"https://huggingface.co/Qwen/Qwen3.8-27B/blob/main/LICENSE","checked":"2026-09-29","sources":"Hugging Face model card · https://huggingface.co/Qwen/Qwen3.8-27B/raw/main/README.md\nHugging Face weights index · https://huggingface.co/Qwen/Qwen3.8-27B/raw/main/model.safetensors.index.json\nGitHub file · https://raw.githubusercontent.com/QwenLM/Qwen3.5/main/README.md (news: 2026-08-14)\nalibabacloud.com announcement · https://www.alibabacloud.com/blog/alibaba-unveils-qwen3-8-27b-and-releases-weights-of-qwen3-8-flagship-model_603463"}},{"slug":"deepseek-v4-pro-0813","name":"DeepSeek-V4-Pro-0813","path":"/open-models/deepseek-v4-pro-0813","category":null,"updatedAt":"2026-09-29T18:43:35.464Z","fields":{"developer":"DeepSeek","released":"August 13, 2026","params":"1.6T","params_b":1600,"active_params":"49B","active_b":49,"architecture":"MoE","context":"1M","context_tokens":1000000,"modalities":"Text","license":"MIT","license_family":"MIT","commercial_use":"Yes","license_terms":"MIT: keep the copyright and permission notice.","precision":"FP4 + FP8","download_gb":null,"official_variants":null,"hf_repo":"https://huggingface.co/deepseek-ai/DeepSeek-V4-Pro-0813","license_url":"https://huggingface.co/deepseek-ai/DeepSeek-V4-Pro-0813/blob/main/LICENSE","checked":"2026-09-29","sources":"Hugging Face model card · https://huggingface.co/deepseek-ai/DeepSeek-V4-Pro-0813/raw/main/README.md\nHugging Face model card · https://huggingface.co/deepseek-ai/DeepSeek-V4-Pro/raw/main/README.md\nHugging Face API · https://huggingface.co/api/models/deepseek-ai/DeepSeek-V4-Pro-0813?expand=safetensors\nDeepSeek API docs · https://api-docs.deepseek.com/updates\nDeepSeek API docs: release notes · https://api-docs.deepseek.com/news/news260424"}},{"slug":"qwen3-8-2-4t-a95b","name":"Qwen3.8-2.4T-A95B","path":"/open-models/qwen3-8-2-4t-a95b","category":null,"updatedAt":"2026-09-29T18:42:59.345Z","fields":{"developer":"Alibaba Qwen","released":"August 12, 2026","params":"2.4T","params_b":2400,"active_params":"95B","active_b":95,"architecture":"MoE","context":"256K","context_tokens":262144,"modalities":"Text","license":"Qwen3.8-Max License","license_family":"Custom license","commercial_use":"Yes, with conditions","license_terms":"Products with more than 100 million monthly active users or more than US$20 million in monthly revenue must display the model name. Model-as-a-service and AI work assistant businesses with more than US$50 million in group revenue over 12 months need a separate license from Qwen. Internal use is exempt.","precision":"BF16","download_gb":4892.4,"official_variants":"Qwen/Qwen3.8-2.4T-A95B-FP8","hf_repo":"https://huggingface.co/Qwen/Qwen3.8-2.4T-A95B","license_url":"https://huggingface.co/Qwen/Qwen3.8-2.4T-A95B/blob/main/LICENSE","checked":"2026-09-29","sources":"Hugging Face model card · https://huggingface.co/Qwen/Qwen3.8-2.4T-A95B/raw/main/README.md\nLicense text on Hugging Face · https://huggingface.co/Qwen/Qwen3.8-2.4T-A95B/raw/main/LICENSE\nHugging Face API · https://huggingface.co/api/models/Qwen/Qwen3.8-2.4T-A95B?expand=safetensors\nGitHub file · https://raw.githubusercontent.com/QwenLM/Qwen3.5/main/README.md (news: 2026-08-12)\nqwen.ai announcement · https://qwen.ai/blog?id=qwen3.8"}},{"slug":"nemotron-3-5-lightning","name":"Nemotron 3.5 Lightning","path":"/open-models/nemotron-3-5-lightning","category":null,"updatedAt":"2026-09-29T18:44:56.192Z","fields":{"developer":"NVIDIA","released":"August 11, 2026","params":"30B","params_b":30,"active_params":"3B","active_b":3,"architecture":"MoE","context":"1M","context_tokens":1000000,"modalities":"Text","license":"OpenMDW License Agreement v1.1","license_family":"Custom license","commercial_use":"Yes","license_terms":"OpenMDW 1.1: redistributions must include the agreement and origin notices.","precision":"BF16","download_gb":65.8,"official_variants":"nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-NVFP4","hf_repo":"https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16","license_url":"https://raw.githubusercontent.com/OpenMDW/OpenMDW/refs/heads/main/1.1/LICENSE.OpenMDW-1.1","checked":"2026-09-29","sources":"Hugging Face model card · https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16/raw/main/README.md\nHugging Face weights index · https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16/raw/main/model.safetensors.index.json"}},{"slug":"kimi-k3","name":"Kimi K3","path":"/open-models/kimi-k3","category":null,"updatedAt":"2026-09-29T18:45:12.545Z","fields":{"developer":"Moonshot AI","released":"July 2026","params":"2.8T","params_b":2800,"active_params":"104B","active_b":104,"architecture":"MoE","context":"1M","context_tokens":1048576,"modalities":"Text, image","license":"Kimi K3 License","license_family":"Custom license","commercial_use":"Yes, with conditions","license_terms":"Model-as-a-service operators with group revenue above US$20 million over 12 months need a separate agreement with Moonshot AI. Products with more than 100 million monthly active users or more than US$20 million in monthly revenue must display \"Kimi K3\". Internal use is exempt.","precision":"MXFP4","download_gb":null,"official_variants":null,"hf_repo":"https://huggingface.co/moonshotai/Kimi-K3","license_url":"https://huggingface.co/moonshotai/Kimi-K3/blob/main/LICENSE","checked":"2026-09-29","sources":"Hugging Face model card · https://huggingface.co/moonshotai/Kimi-K3/raw/main/README.md\nLicense text on Hugging Face · https://huggingface.co/moonshotai/Kimi-K3/raw/main/LICENSE\nkimi.com announcement · https://www.kimi.com/blog/kimi-k3"}},{"slug":"nemotron-3-ultra","name":"Nemotron 3 Ultra","path":"/open-models/nemotron-3-ultra","category":null,"updatedAt":"2026-09-29T18:44:44.225Z","fields":{"developer":"NVIDIA","released":"June 4, 2026","params":"550B","params_b":550,"active_params":"55B","active_b":55,"architecture":"MoE","context":"1M","context_tokens":1000000,"modalities":"Text","license":"OpenMDW License Agreement v1.1","license_family":"Custom license","commercial_use":"Yes","license_terms":"OpenMDW 1.1: redistributions must include the agreement and origin notices, and the rights end for anyone who sues claiming the materials infringe.","precision":"BF16","download_gb":1121,"official_variants":"nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4","hf_repo":"https://huggingface.co/nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16","license_url":"https://raw.githubusercontent.com/OpenMDW/OpenMDW/refs/heads/main/1.1/LICENSE.OpenMDW-1.1","checked":"2026-09-29","sources":"Hugging Face model card · https://huggingface.co/nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16/raw/main/README.md\nLicense text on GitHub · https://raw.githubusercontent.com/OpenMDW/OpenMDW/refs/heads/main/1.1/LICENSE.OpenMDW-1.1\nHugging Face weights index · https://huggingface.co/nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16/raw/main/model.safetensors.index.json"}},{"slug":"gemma-4-12b","name":"Gemma 4 12B","path":"/open-models/gemma-4-12b","category":null,"updatedAt":"2026-09-29T18:42:21.350Z","fields":{"developer":"Google DeepMind","released":"June 3, 2026","params":"11.95B","params_b":11.95,"active_params":null,"active_b":null,"architecture":"Dense","context":"256K","context_tokens":256000,"modalities":"Text, image, audio","license":"Apache 2.0","license_family":"Apache 2.0","commercial_use":"Yes","license_terms":"Apache 2.0: keep the license and notices; no limits on field of use or number of users. Gemma 4 is the first Gemma generation under Apache 2.0; earlier versions used the Gemma Terms of Use.","precision":"BF16","download_gb":23.9,"official_variants":"google/gemma-4-12B-it-qat-q4_0-gguf; google/gemma-4-12B-it-qat-q4_0-unquantized; google/gemma-4-12B-it-qat-w4a16-ct","hf_repo":"https://huggingface.co/google/gemma-4-12B-it","license_url":"https://ai.google.dev/gemma/docs/gemma_4_license","checked":"2026-09-29","sources":"Hugging Face model card · https://huggingface.co/google/gemma-4-31B-it/raw/main/README.md\nblog.google announcement · https://blog.google/innovation-and-ai/technology/developers-tools/gemma-4/\nopensource.googleblog.com announcement · https://opensource.googleblog.com/2026/03/gemma-4-expanding-the-gemmaverse-with-apache-20.html\nHugging Face API · https://huggingface.co/api/models?author=google&search=gemma-4\nblog.google announcement · https://blog.google/innovation-and-ai/technology/developers-tools/introducing-gemma-4-12b/"}},{"slug":"minimax-m3","name":"MiniMax-M3","path":"/open-models/minimax-m3","category":null,"updatedAt":"2026-09-29T18:45:38.178Z","fields":{"developer":"MiniMax","released":"May 31, 2026","params":"~428B","params_b":428,"active_params":"~23B","active_b":23,"architecture":"MoE","context":"1M","context_tokens":1000000,"modalities":"Text, image, video","license":"MiniMax Community License","license_family":"Custom license","commercial_use":"Yes, with conditions","license_terms":"Commercial users must display \"Built with MiniMax M3\" and either notify MiniMax once (up to US$20 million in yearly revenue) or get prior written authorization (above US$20 million). The license lists prohibited uses, including military use.","precision":"BF16","download_gb":869.2,"official_variants":"MiniMaxAI/MiniMax-M3-MXFP8","hf_repo":"https://huggingface.co/MiniMaxAI/MiniMax-M3","license_url":"https://huggingface.co/MiniMaxAI/MiniMax-M3/blob/main/LICENSE","checked":"2026-09-29","sources":"Hugging Face model card · https://huggingface.co/MiniMaxAI/MiniMax-M3/raw/main/README.md\nLicense text on Hugging Face · https://huggingface.co/MiniMaxAI/MiniMax-M3/raw/main/LICENSE\nHugging Face API · https://huggingface.co/api/models/MiniMaxAI/MiniMax-M3?expand=safetensors\nminimax.io announcement · https://www.minimax.io/blog/minimax-m3"}},{"slug":"mistral-medium-3-5","name":"Mistral Medium 3.5","path":"/open-models/mistral-medium-3-5","category":null,"updatedAt":"2026-09-29T18:44:00.090Z","fields":{"developer":"Mistral AI","released":"May 22, 2026","params":"128B","params_b":128,"active_params":null,"active_b":null,"architecture":"Dense","context":"256K","context_tokens":256000,"modalities":"Text, image","license":"Modified MIT License","license_family":"Modified MIT","commercial_use":"Yes, with conditions","license_terms":"Companies with global consolidated monthly revenue above US$20 million, or their employees, may not use the model without a separate commercial license from Mistral AI. An attribution notice is required.","precision":"FP8 + BF16","download_gb":133.6,"official_variants":"mistralai/Mistral-Medium-3.5-128B-EAGLE (speculative-decoding head)","hf_repo":"https://huggingface.co/mistralai/Mistral-Medium-3.5-128B","license_url":"https://huggingface.co/mistralai/Mistral-Medium-3.5-128B/blob/main/LICENSE","checked":"2026-09-29","sources":"Hugging Face model card · https://huggingface.co/mistralai/Mistral-Medium-3.5-128B/raw/main/README.md\nLicense text on Hugging Face · https://huggingface.co/mistralai/Mistral-Medium-3.5-128B/raw/main/LICENSE\nHugging Face API · https://huggingface.co/api/models/mistralai/Mistral-Medium-3.5-128B?expand=safetensors\nmistral.ai announcement · https://mistral.ai/news/vibe-remote-agents-mistral-medium-3-5"}},{"slug":"command-a-plus","name":"Command A+","path":"/open-models/command-a-plus","category":null,"updatedAt":"2026-09-29T18:45:55.055Z","fields":{"developer":"Cohere","released":"May 20, 2026","params":"218B","params_b":218,"active_params":"25B","active_b":25,"architecture":"MoE","context":"128K","context_tokens":128000,"modalities":"Text, image","license":"Apache 2.0","license_family":"Apache 2.0","commercial_use":"Yes","license_terms":"Apache 2.0, unlike earlier Cohere open-weight releases, which were non-commercial.","precision":"BF16","download_gb":437.5,"official_variants":"CohereLabs/command-a-plus-05-2026-fp8; CohereLabs/command-a-plus-05-2026-w4a4","hf_repo":"https://huggingface.co/CohereLabs/command-a-plus-05-2026-bf16","license_url":"https://www.apache.org/licenses/LICENSE-2.0","checked":"2026-09-29","sources":"Hugging Face model page · https://huggingface.co/CohereLabs/command-a-plus-05-2026-bf16\ncohere.com announcement · https://cohere.com/blog/command-a-plus\nHugging Face API · https://huggingface.co/api/models/CohereLabs/command-a-plus-05-2026-bf16?expand=safetensors"}},{"slug":"kimi-k2-6","name":"Kimi K2.6","path":"/open-models/kimi-k2-6","category":null,"updatedAt":"2026-09-29T18:45:18.323Z","fields":{"developer":"Moonshot AI","released":"April 21, 2026","params":"1T","params_b":1000,"active_params":"32B","active_b":32,"architecture":"MoE","context":"256K","context_tokens":256000,"modalities":"Text, image","license":"Modified MIT License","license_family":"Modified MIT","commercial_use":"Yes, with conditions","license_terms":"Commercial products with more than 100 million monthly active users or more than US$20 million in monthly revenue must display \"Kimi K2.6\" in their interface.","precision":null,"download_gb":null,"official_variants":null,"hf_repo":"https://huggingface.co/moonshotai/Kimi-K2.6","license_url":"https://huggingface.co/moonshotai/Kimi-K2.6/blob/main/LICENSE","checked":"2026-09-29","sources":"Hugging Face model card · https://huggingface.co/moonshotai/Kimi-K2.6/raw/main/README.md\nLicense text on Hugging Face · https://huggingface.co/moonshotai/Kimi-K2.6/raw/main/LICENSE\nforum.moonshot.ai · https://forum.moonshot.ai/t/meet-kimi-k2-6-advancing-open-source-coding/369"}},{"slug":"qwen3-6-35b-a3b","name":"Qwen3.6-35B-A3B","path":"/open-models/qwen3-6-35b-a3b","category":null,"updatedAt":"2026-09-29T18:43:12.191Z","fields":{"developer":"Alibaba Qwen","released":"April 16, 2026","params":"35B","params_b":35,"active_params":"3B","active_b":3,"architecture":"MoE","context":"256K","context_tokens":262144,"modalities":"Text, image","license":"Apache 2.0","license_family":"Apache 2.0","commercial_use":"Yes","license_terms":"Apache 2.0, with no user or revenue limits.","precision":null,"download_gb":71.9,"official_variants":"Qwen/Qwen3.6-35B-A3B-FP8","hf_repo":"https://huggingface.co/Qwen/Qwen3.6-35B-A3B","license_url":"https://huggingface.co/Qwen/Qwen3.6-35B-A3B/blob/main/LICENSE","checked":"2026-09-29","sources":"Hugging Face model card · https://huggingface.co/Qwen/Qwen3.6-35B-A3B/raw/main/README.md\nHugging Face weights index · https://huggingface.co/Qwen/Qwen3.6-35B-A3B/raw/main/model.safetensors.index.json\nGitHub file · https://raw.githubusercontent.com/QwenLM/Qwen3.6/main/README.md"}},{"slug":"gemma-4-26b-a4b","name":"Gemma 4 26B A4B","path":"/open-models/gemma-4-26b-a4b","category":null,"updatedAt":"2026-09-29T18:42:15.650Z","fields":{"developer":"Google DeepMind","released":"April 2, 2026","params":"25.2B","params_b":25.2,"active_params":"3.8B","active_b":3.8,"architecture":"MoE","context":"256K","context_tokens":256000,"modalities":"Text, image","license":"Apache 2.0","license_family":"Apache 2.0","commercial_use":"Yes","license_terms":"Apache 2.0: keep the license and notices; no limits on field of use or number of users. Gemma 4 is the first Gemma generation under Apache 2.0; earlier versions used the Gemma Terms of Use.","precision":"BF16","download_gb":51.6,"official_variants":"google/gemma-4-26B-A4B-it-qat-q4_0-gguf; google/gemma-4-26B-A4B-it-qat-q4_0-unquantized","hf_repo":"https://huggingface.co/google/gemma-4-26B-A4B-it","license_url":"https://ai.google.dev/gemma/docs/gemma_4_license","checked":"2026-09-29","sources":"Hugging Face model card · https://huggingface.co/google/gemma-4-31B-it/raw/main/README.md\nblog.google announcement · https://blog.google/innovation-and-ai/technology/developers-tools/gemma-4/\nopensource.googleblog.com announcement · https://opensource.googleblog.com/2026/03/gemma-4-expanding-the-gemmaverse-with-apache-20.html\nHugging Face API · https://huggingface.co/api/models?author=google&search=gemma-4\nHugging Face weights index · https://huggingface.co/google/gemma-4-26B-A4B-it/raw/main/model.safetensors.index.json"}},{"slug":"gemma-4-31b","name":"Gemma 4 31B","path":"/open-models/gemma-4-31b","category":null,"updatedAt":"2026-09-29T18:42:09.513Z","fields":{"developer":"Google DeepMind","released":"April 2, 2026","params":"30.7B","params_b":30.7,"active_params":null,"active_b":null,"architecture":"Dense","context":"256K","context_tokens":256000,"modalities":"Text, image","license":"Apache 2.0","license_family":"Apache 2.0","commercial_use":"Yes","license_terms":"Apache 2.0: keep the license and notices; no limits on field of use or number of users. Gemma 4 is the first Gemma generation under Apache 2.0; earlier versions used the Gemma Terms of Use.","precision":"BF16","download_gb":62.5,"official_variants":"google/gemma-4-31B-it-qat-q4_0-gguf; google/gemma-4-31B-it-qat-q4_0-unquantized; google/gemma-4-31B-it-qat-w4a16-ct","hf_repo":"https://huggingface.co/google/gemma-4-31B-it","license_url":"https://ai.google.dev/gemma/docs/gemma_4_license","checked":"2026-09-29","sources":"Hugging Face model card · https://huggingface.co/google/gemma-4-31B-it/raw/main/README.md\nblog.google announcement · https://blog.google/innovation-and-ai/technology/developers-tools/gemma-4/\nopensource.googleblog.com announcement · https://opensource.googleblog.com/2026/03/gemma-4-expanding-the-gemmaverse-with-apache-20.html\nHugging Face API · https://huggingface.co/api/models?author=google&search=gemma-4\nHugging Face weights index · https://huggingface.co/google/gemma-4-31B-it/raw/main/model.safetensors.index.json"}},{"slug":"gemma-4-e2b","name":"Gemma 4 E2B","path":"/open-models/gemma-4-e2b","category":null,"updatedAt":"2026-09-29T18:42:33.734Z","fields":{"developer":"Google DeepMind","released":"April 2, 2026","params":"5.1B","params_b":5.1,"active_params":null,"active_b":null,"architecture":"Dense","context":"128K","context_tokens":128000,"modalities":"Text, image, audio","license":"Apache 2.0","license_family":"Apache 2.0","commercial_use":"Yes","license_terms":"Apache 2.0: keep the license and notices; no limits on field of use or number of users. Gemma 4 is the first Gemma generation under Apache 2.0; earlier versions used the Gemma Terms of Use.","precision":"BF16","download_gb":10.2,"official_variants":"google/gemma-4-E2B-it-qat-q4_0-gguf; google/gemma-4-E2B-it-qat-q4_0-unquantized; google/gemma-4-E2B-it-qat-w4a16-ct; google/gemma-4-E2B-it-qat-mobile-ct; google/gemma-4-E2B-it-qat-mobile-transformers","hf_repo":"https://huggingface.co/google/gemma-4-E2B-it","license_url":"https://ai.google.dev/gemma/docs/gemma_4_license","checked":"2026-09-29","sources":"Hugging Face model card · https://huggingface.co/google/gemma-4-31B-it/raw/main/README.md\nblog.google announcement · https://blog.google/innovation-and-ai/technology/developers-tools/gemma-4/\nopensource.googleblog.com announcement · https://opensource.googleblog.com/2026/03/gemma-4-expanding-the-gemmaverse-with-apache-20.html\nHugging Face API · https://huggingface.co/api/models?author=google&search=gemma-4"}},{"slug":"gemma-4-e4b","name":"Gemma 4 E4B","path":"/open-models/gemma-4-e4b","category":null,"updatedAt":"2026-09-29T18:42:28.387Z","fields":{"developer":"Google DeepMind","released":"April 2, 2026","params":"8B","params_b":8,"active_params":null,"active_b":null,"architecture":"Dense","context":"128K","context_tokens":128000,"modalities":"Text, image, audio","license":"Apache 2.0","license_family":"Apache 2.0","commercial_use":"Yes","license_terms":"Apache 2.0: keep the license and notices; no limits on field of use or number of users. Gemma 4 is the first Gemma generation under Apache 2.0; earlier versions used the Gemma Terms of Use.","precision":"BF16","download_gb":16,"official_variants":"google/gemma-4-E4B-it-qat-q4_0-gguf; google/gemma-4-E4B-it-qat-q4_0-unquantized; google/gemma-4-E4B-it-qat-w4a16-ct; google/gemma-4-E4B-it-qat-mobile-ct; google/gemma-4-E4B-it-qat-mobile-transformers","hf_repo":"https://huggingface.co/google/gemma-4-E4B-it","license_url":"https://ai.google.dev/gemma/docs/gemma_4_license","checked":"2026-09-29","sources":"Hugging Face model card · https://huggingface.co/google/gemma-4-31B-it/raw/main/README.md\nblog.google announcement · https://blog.google/innovation-and-ai/technology/developers-tools/gemma-4/\nopensource.googleblog.com announcement · https://opensource.googleblog.com/2026/03/gemma-4-expanding-the-gemmaverse-with-apache-20.html\nHugging Face API · https://huggingface.co/api/models?author=google&search=gemma-4"}},{"slug":"minimax-m2-7","name":"MiniMax-M2.7","path":"/open-models/minimax-m2-7","category":null,"updatedAt":"2026-09-29T18:45:43.743Z","fields":{"developer":"MiniMax","released":"March 18, 2026","params":"229B","params_b":229,"active_params":null,"active_b":null,"architecture":"MoE","context":null,"context_tokens":null,"modalities":"Text","license":"MiniMax Non-Commercial License","license_family":"Non-commercial","commercial_use":"No","license_terms":"Personal, non-profit and academic research use only. Any commercial use needs prior written authorization from MiniMax and the attribution \"Built with MiniMax M2.7\".","precision":"FP8 + BF16","download_gb":null,"official_variants":null,"hf_repo":"https://huggingface.co/MiniMaxAI/MiniMax-M2.7","license_url":"https://github.com/MiniMax-AI/MiniMax-M2.7/blob/main/LICENSE","checked":"2026-09-29","sources":"Hugging Face model card · https://huggingface.co/MiniMaxAI/MiniMax-M2.7/raw/main/README.md\nLicense text on GitHub · https://raw.githubusercontent.com/MiniMax-AI/MiniMax-M2.7/main/LICENSE\nHugging Face API · https://huggingface.co/api/models/MiniMaxAI/MiniMax-M2.7?expand=safetensors\nminimax.io announcement · https://www.minimax.io/news/minimax-m27-en"}},{"slug":"mistral-small-4","name":"Mistral Small 4","path":"/open-models/mistral-small-4","category":null,"updatedAt":"2026-09-29T18:43:53.826Z","fields":{"developer":"Mistral AI","released":"March 16, 2026","params":"119B","params_b":119,"active_params":"6.5B","active_b":6.5,"architecture":"MoE","context":"256K","context_tokens":256000,"modalities":"Text, image","license":"Apache 2.0","license_family":"Apache 2.0","commercial_use":"Yes","license_terms":"Apache 2.0, with no user or revenue limits.","precision":"FP8 + BF16","download_gb":120.9,"official_variants":"mistralai/Mistral-Small-4-119B-2603-NVFP4; mistralai/Mistral-Small-4-119B-2603-eagle (speculative-decoding head)","hf_repo":"https://huggingface.co/mistralai/Mistral-Small-4-119B-2603","license_url":"https://www.apache.org/licenses/LICENSE-2.0","checked":"2026-09-29","sources":"Hugging Face model card · https://huggingface.co/mistralai/Mistral-Small-4-119B-2603/raw/main/README.md\nHugging Face weights index · https://huggingface.co/mistralai/Mistral-Small-4-119B-2603/raw/main/model.safetensors.index.json\nHugging Face API · https://huggingface.co/api/models/mistralai/Mistral-Small-4-119B-2603?expand=safetensors\nmistral.ai announcement · https://mistral.ai/news/mistral-small-4/"}},{"slug":"nemotron-3-super","name":"Nemotron 3 Super","path":"/open-models/nemotron-3-super","category":null,"updatedAt":"2026-09-29T18:44:50.647Z","fields":{"developer":"NVIDIA","released":"March 11, 2026","params":"120B","params_b":120,"active_params":"12B","active_b":12,"architecture":"MoE","context":"1M","context_tokens":1000000,"modalities":"Text","license":"NVIDIA Nemotron Open Model License","license_family":"Custom license","commercial_use":"Yes","license_terms":"Commercial use is allowed. Redistributions must include the license and the attribution \"Licensed by NVIDIA Corporation under the NVIDIA Nemotron Model License\"; patent or copyright litigation ends the license, and export controls apply.","precision":"BF16","download_gb":247.2,"official_variants":"nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-FP8; nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4","hf_repo":"https://huggingface.co/nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16","license_url":"https://www.nvidia.com/en-us/agreements/enterprise-software/nvidia-nemotron-open-model-license/","checked":"2026-09-29","sources":"Hugging Face model card · https://huggingface.co/nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16/raw/main/README.md\nLicense text (nvidia.com) · https://www.nvidia.com/en-us/agreements/enterprise-software/nvidia-nemotron-open-model-license/\nHugging Face weights index · https://huggingface.co/nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16/raw/main/model.safetensors.index.json"}},{"slug":"phi-4-reasoning-vision-15b","name":"Phi-4-reasoning-vision-15B","path":"/open-models/phi-4-reasoning-vision-15b","category":null,"updatedAt":"2026-09-29T18:44:31.424Z","fields":{"developer":"Microsoft","released":"March 4, 2026","params":"15B","params_b":15,"active_params":null,"active_b":null,"architecture":"Dense","context":"16K","context_tokens":16384,"modalities":"Text, image","license":"MIT","license_family":"MIT","commercial_use":"Yes","license_terms":"MIT: keep the copyright and permission notice.","precision":"BF16","download_gb":30.2,"official_variants":null,"hf_repo":"https://huggingface.co/microsoft/Phi-4-reasoning-vision-15B","license_url":"https://opensource.org/licenses/MIT","checked":"2026-09-29","sources":"Hugging Face model card · https://huggingface.co/microsoft/Phi-4-reasoning-vision-15B/raw/main/README.md\nHugging Face API · https://huggingface.co/api/models/microsoft/Phi-4-reasoning-vision-15B?expand=safetensors"}},{"slug":"qwen3-5-397b-a17b","name":"Qwen3.5-397B-A17B","path":"/open-models/qwen3-5-397b-a17b","category":null,"updatedAt":"2026-09-29T18:43:18.089Z","fields":{"developer":"Alibaba Qwen","released":"February 16, 2026","params":"397B","params_b":397,"active_params":"17B","active_b":17,"architecture":"MoE","context":"256K","context_tokens":262144,"modalities":"Text, image","license":"Apache 2.0","license_family":"Apache 2.0","commercial_use":"Yes","license_terms":"Apache 2.0, with no user or revenue limits.","precision":null,"download_gb":806.8,"official_variants":"Qwen/Qwen3.5-397B-A17B-FP8; Qwen/Qwen3.5-397B-A17B-GPTQ-Int4","hf_repo":"https://huggingface.co/Qwen/Qwen3.5-397B-A17B","license_url":"https://huggingface.co/Qwen/Qwen3.5-397B-A17B/blob/main/LICENSE","checked":"2026-09-29","sources":"Hugging Face model card · https://huggingface.co/Qwen/Qwen3.5-397B-A17B/raw/main/README.md\nHugging Face weights index · https://huggingface.co/Qwen/Qwen3.5-397B-A17B/raw/main/model.safetensors.index.json\nGitHub file · https://raw.githubusercontent.com/QwenLM/Qwen3.6/main/README.md"}},{"slug":"qwen3-coder-next","name":"Qwen3-Coder-Next","path":"/open-models/qwen3-coder-next","category":null,"updatedAt":"2026-09-29T18:43:23.947Z","fields":{"developer":"Alibaba Qwen","released":"February 3, 2026","params":"80B","params_b":80,"active_params":"3B","active_b":3,"architecture":"MoE","context":"256K","context_tokens":262144,"modalities":"Text","license":"Apache 2.0","license_family":"Apache 2.0","commercial_use":"Yes","license_terms":"Apache 2.0, with no user or revenue limits.","precision":null,"download_gb":159.3,"official_variants":"Qwen/Qwen3-Coder-Next-FP8; Qwen/Qwen3-Coder-Next-GGUF","hf_repo":"https://huggingface.co/Qwen/Qwen3-Coder-Next","license_url":"https://huggingface.co/Qwen/Qwen3-Coder-Next/blob/main/LICENSE","checked":"2026-09-29","sources":"Hugging Face model card · https://huggingface.co/Qwen/Qwen3-Coder-Next/raw/main/README.md\nHugging Face weights index · https://huggingface.co/Qwen/Qwen3-Coder-Next/raw/main/model.safetensors.index.json\nGitHub file · https://raw.githubusercontent.com/QwenLM/Qwen3-Coder/main/README.md\nQwen on LinkedIn · https://www.linkedin.com/posts/qwen_introducingqwen3-coder-next-an-open-weight-activity-7424484355294576640-1Vi0"}},{"slug":"olmo-3-1-32b","name":"Olmo 3.1 32B","path":"/open-models/olmo-3-1-32b","category":null,"updatedAt":"2026-09-29T18:45:49.325Z","fields":{"developer":"Allen Institute for AI (Ai2)","released":"December 12, 2025","params":"32B","params_b":32,"active_params":null,"active_b":null,"architecture":"Dense","context":"65K","context_tokens":null,"modalities":"Text","license":"Apache 2.0","license_family":"Apache 2.0","commercial_use":"Yes","license_terms":"Apache 2.0. Ai2 also publishes Responsible Use Guidelines, which are guidance rather than license terms.","precision":"BF16","download_gb":64.5,"official_variants":null,"hf_repo":"https://huggingface.co/allenai/Olmo-3.1-32B-Instruct","license_url":"https://www.apache.org/licenses/LICENSE-2.0","checked":"2026-09-29","sources":"Hugging Face model card · https://huggingface.co/allenai/Olmo-3.1-32B-Instruct/raw/main/README.md\nHugging Face model page · https://huggingface.co/allenai/Olmo-3.1-32B-Instruct\nHugging Face API · https://huggingface.co/api/models/allenai/Olmo-3.1-32B-Instruct?expand=safetensors\nallenai.org announcement · https://allenai.org/blog/olmo3"}},{"slug":"devstral-2","name":"Devstral 2","path":"/open-models/devstral-2","category":null,"updatedAt":"2026-09-29T18:44:10.569Z","fields":{"developer":"Mistral AI","released":"December 9, 2025","params":"123B","params_b":123,"active_params":null,"active_b":null,"architecture":"Dense","context":"256K","context_tokens":256000,"modalities":"Text","license":"Modified MIT License","license_family":"Modified MIT","commercial_use":"Yes, with conditions","license_terms":"Companies with global consolidated monthly revenue above US$20 million need a separate commercial license from Mistral AI.","precision":"FP8","download_gb":128.2,"official_variants":null,"hf_repo":"https://huggingface.co/mistralai/Devstral-2-123B-Instruct-2512","license_url":"https://huggingface.co/mistralai/Devstral-2-123B-Instruct-2512/blob/main/LICENSE","checked":"2026-09-29","sources":"Hugging Face model card · https://huggingface.co/mistralai/Devstral-2-123B-Instruct-2512/raw/main/README.md\nLicense text on Hugging Face · https://huggingface.co/mistralai/Devstral-2-123B-Instruct-2512/raw/main/LICENSE\nHugging Face weights index · https://huggingface.co/mistralai/Devstral-2-123B-Instruct-2512/raw/main/model.safetensors.index.json\nmistral.ai announcement · https://mistral.ai/news/devstral-2-vibe-cli"}},{"slug":"devstral-small-2","name":"Devstral Small 2","path":"/open-models/devstral-small-2","category":null,"updatedAt":"2026-09-29T18:44:16.016Z","fields":{"developer":"Mistral AI","released":"December 9, 2025","params":"24B","params_b":24,"active_params":null,"active_b":null,"architecture":"Dense","context":"256K","context_tokens":256000,"modalities":"Text, image","license":"Apache 2.0","license_family":"Apache 2.0","commercial_use":"Yes","license_terms":"Apache 2.0, with no user or revenue limits.","precision":"FP8","download_gb":25.8,"official_variants":null,"hf_repo":"https://huggingface.co/mistralai/Devstral-Small-2-24B-Instruct-2512","license_url":"https://www.apache.org/licenses/LICENSE-2.0","checked":"2026-09-29","sources":"Hugging Face model card · https://huggingface.co/mistralai/Devstral-Small-2-24B-Instruct-2512/raw/main/README.md\nHugging Face weights index · https://huggingface.co/mistralai/Devstral-Small-2-24B-Instruct-2512/raw/main/model.safetensors.index.json\nmistral.ai announcement · https://mistral.ai/news/devstral-2-vibe-cli"}},{"slug":"ministral-3-14b","name":"Ministral 3 14B","path":"/open-models/ministral-3-14b","category":null,"updatedAt":"2026-09-29T18:44:21.313Z","fields":{"developer":"Mistral AI","released":"December 2, 2025","params":"14B","params_b":14,"active_params":null,"active_b":null,"architecture":"Dense","context":"256K","context_tokens":256000,"modalities":"Text, image","license":"Apache 2.0","license_family":"Apache 2.0","commercial_use":"Yes","license_terms":"Apache 2.0, with no user or revenue limits.","precision":"FP8","download_gb":15.7,"official_variants":"mistralai/Ministral-3-14B-Instruct-2512-GGUF; mistralai/Ministral-3-14B-Instruct-2512-BF16 (full precision)","hf_repo":"https://huggingface.co/mistralai/Ministral-3-14B-Instruct-2512","license_url":"https://www.apache.org/licenses/LICENSE-2.0","checked":"2026-09-29","sources":"Hugging Face model card · https://huggingface.co/mistralai/Ministral-3-14B-Instruct-2512/raw/main/README.md\nHugging Face weights index · https://huggingface.co/mistralai/Ministral-3-14B-Instruct-2512/raw/main/model.safetensors.index.json\nmistral.ai announcement · https://mistral.ai/news/mistral-3"}},{"slug":"mistral-large-3","name":"Mistral Large 3","path":"/open-models/mistral-large-3","category":null,"updatedAt":"2026-09-29T18:44:05.200Z","fields":{"developer":"Mistral AI","released":"December 2, 2025","params":"675B","params_b":675,"active_params":"41B","active_b":41,"architecture":"MoE","context":"256K","context_tokens":256000,"modalities":"Text, image","license":"Apache 2.0","license_family":"Apache 2.0","commercial_use":"Yes","license_terms":"Apache 2.0, with no user or revenue limits.","precision":"FP8","download_gb":null,"official_variants":"mistralai/Mistral-Large-3-675B-Instruct-2512-NVFP4; mistralai/Mistral-Large-3-675B-Instruct-2512-BF16 (full precision); mistralai/Mistral-Large-3-675B-Instruct-2512-Eagle","hf_repo":"https://huggingface.co/mistralai/Mistral-Large-3-675B-Instruct-2512","license_url":"https://www.apache.org/licenses/LICENSE-2.0","checked":"2026-09-29","sources":"Hugging Face model card · https://huggingface.co/mistralai/Mistral-Large-3-675B-Instruct-2512/raw/main/README.md\nmistral.ai announcement · https://mistral.ai/news/mistral-3"}},{"slug":"deepseek-v3-2","name":"DeepSeek-V3.2","path":"/open-models/deepseek-v3-2","category":null,"updatedAt":"2026-09-29T18:43:42.416Z","fields":{"developer":"DeepSeek","released":"December 1, 2025","params":"671B","params_b":671,"active_params":"37B","active_b":37,"architecture":"MoE","context":null,"context_tokens":null,"modalities":"Text","license":"MIT","license_family":"MIT","commercial_use":"Yes","license_terms":"MIT: keep the copyright and permission notice.","precision":"FP8 + BF16","download_gb":1370.8,"official_variants":null,"hf_repo":"https://huggingface.co/deepseek-ai/DeepSeek-V3.2","license_url":"https://huggingface.co/deepseek-ai/DeepSeek-V3.2/blob/main/LICENSE","checked":"2026-09-29","sources":"Hugging Face model card · https://huggingface.co/deepseek-ai/DeepSeek-V3.2/raw/main/README.md\nHugging Face weights index · https://huggingface.co/deepseek-ai/DeepSeek-V3.2/raw/main/model.safetensors.index.json\nDeepSeek API docs: release notes · https://api-docs.deepseek.com/news/news251201\nHugging Face model card · https://huggingface.co/deepseek-ai/DeepSeek-V4-Pro/raw/main/README.md"}},{"slug":"magistral-small-1-2","name":"Magistral Small 1.2","path":"/open-models/magistral-small-1-2","category":null,"updatedAt":"2026-09-29T18:44:26.176Z","fields":{"developer":"Mistral AI","released":"September 2025","params":"24B","params_b":24,"active_params":null,"active_b":null,"architecture":"Dense","context":"128K","context_tokens":128000,"modalities":"Text, image","license":"Apache 2.0","license_family":"Apache 2.0","commercial_use":"Yes","license_terms":"Apache 2.0, with no user or revenue limits.","precision":"BF16","download_gb":48,"official_variants":"mistralai/Magistral-Small-2509-GGUF","hf_repo":"https://huggingface.co/mistralai/Magistral-Small-2509","license_url":"https://www.apache.org/licenses/LICENSE-2.0","checked":"2026-09-29","sources":"Hugging Face model card · https://huggingface.co/mistralai/Magistral-Small-2509/raw/main/README.md\nHugging Face weights index · https://huggingface.co/mistralai/Magistral-Small-2509/raw/main/model.safetensors.index.json"}},{"slug":"gpt-oss-20b","name":"gpt-oss-20b","path":"/open-models/gpt-oss-20b","category":null,"updatedAt":"2026-09-29T18:42:46.897Z","fields":{"developer":"OpenAI","released":"August 5, 2025","params":"21B","params_b":21,"active_params":"3.6B","active_b":3.6,"architecture":"MoE","context":"128K","context_tokens":128000,"modalities":"Text","license":"Apache 2.0","license_family":"Apache 2.0","commercial_use":"Yes","license_terms":"Apache 2.0, with a short usage policy in the repository. No user or revenue limits.","precision":"MXFP4 + BF16","download_gb":13.8,"official_variants":null,"hf_repo":"https://huggingface.co/openai/gpt-oss-20b","license_url":"https://huggingface.co/openai/gpt-oss-20b/blob/main/LICENSE","checked":"2026-09-29","sources":"Hugging Face model card · https://huggingface.co/openai/gpt-oss-20b/raw/main/README.md\nHugging Face weights index · https://huggingface.co/openai/gpt-oss-20b/raw/main/model.safetensors.index.json\nopenai.com · https://openai.com/index/introducing-gpt-oss/\nHugging Face API · https://huggingface.co/api/models/openai/gpt-oss-20b?expand=safetensors"}},{"slug":"gpt-oss-120b","name":"gpt-oss-120b","path":"/open-models/gpt-oss-120b","category":null,"updatedAt":"2026-09-29T18:42:40.794Z","fields":{"developer":"OpenAI","released":"August 5, 2025","params":"117B","params_b":117,"active_params":"5.1B","active_b":5.1,"architecture":"MoE","context":"128K","context_tokens":128000,"modalities":"Text","license":"Apache 2.0","license_family":"Apache 2.0","commercial_use":"Yes","license_terms":"Apache 2.0, with a short usage policy in the repository. No user or revenue limits.","precision":"MXFP4 + BF16","download_gb":65.2,"official_variants":null,"hf_repo":"https://huggingface.co/openai/gpt-oss-120b","license_url":"https://huggingface.co/openai/gpt-oss-120b/blob/main/LICENSE","checked":"2026-09-29","sources":"Hugging Face model card · https://huggingface.co/openai/gpt-oss-20b/raw/main/README.md\nHugging Face weights index · https://huggingface.co/openai/gpt-oss-120b/raw/main/model.safetensors.index.json\nopenai.com · https://openai.com/index/introducing-gpt-oss/\nHugging Face API · https://huggingface.co/api/models/openai/gpt-oss-120b?expand=safetensors"}},{"slug":"deepseek-r1-0528","name":"DeepSeek-R1-0528","path":"/open-models/deepseek-r1-0528","category":null,"updatedAt":"2026-09-29T18:43:47.941Z","fields":{"developer":"DeepSeek","released":"May 28, 2025","params":"671B","params_b":671,"active_params":"37B","active_b":37,"architecture":"MoE","context":"128K","context_tokens":128000,"modalities":"Text","license":"MIT","license_family":"MIT","commercial_use":"Yes","license_terms":"MIT: keep the copyright and permission notice.","precision":"FP8 + BF16","download_gb":1369.1,"official_variants":null,"hf_repo":"https://huggingface.co/deepseek-ai/DeepSeek-R1-0528","license_url":"https://huggingface.co/deepseek-ai/DeepSeek-R1-0528/blob/main/LICENSE","checked":"2026-09-29","sources":"Hugging Face model card · https://huggingface.co/deepseek-ai/DeepSeek-R1-0528/raw/main/README.md\nHugging Face weights index · https://huggingface.co/deepseek-ai/DeepSeek-R1-0528/raw/main/model.safetensors.index.json\nHugging Face model page · https://huggingface.co/deepseek-ai/DeepSeek-R1\nDeepSeek API docs: release notes · https://api-docs.deepseek.com/news/news250528"}},{"slug":"llama-4-maverick","name":"Llama 4 Maverick","path":"/open-models/llama-4-maverick","category":null,"updatedAt":"2026-09-29T18:41:47.628Z","fields":{"developer":"Meta","released":"April 5, 2025","params":"400B","params_b":400,"active_params":"17B","active_b":17,"architecture":"MoE","context":"1M","context_tokens":1000000,"modalities":"Text, image","license":"Llama 4 Community License","license_family":"Custom license","commercial_use":"Yes, with conditions","license_terms":"Companies whose products had more than 700 million monthly active users on the release date need a separate license from Meta. Products must display \"Built with Llama\", model names derived from it must start with \"Llama\", and the Llama Acceptable Use Policy applies.","precision":"BF16","download_gb":803.2,"official_variants":"meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8","hf_repo":"https://huggingface.co/meta-llama/Llama-4-Maverick-17B-128E-Instruct","license_url":"https://github.com/meta-llama/llama-models/blob/main/models/llama4/LICENSE","checked":"2026-09-29","sources":"Hugging Face model page · https://huggingface.co/meta-llama/Llama-4-Scout-17B-16E-Instruct (shared Llama 4 card)\nHugging Face API · https://huggingface.co/api/models/meta-llama/Llama-4-Maverick-17B-128E-Instruct?expand=safetensors\nHugging Face API · https://huggingface.co/api/models?author=meta-llama&search=FP8"}},{"slug":"llama-4-scout","name":"Llama 4 Scout","path":"/open-models/llama-4-scout","category":null,"updatedAt":"2026-09-29T18:41:41.371Z","fields":{"developer":"Meta","released":"April 5, 2025","params":"109B","params_b":109,"active_params":"17B","active_b":17,"architecture":"MoE","context":"10M","context_tokens":10000000,"modalities":"Text, image","license":"Llama 4 Community License","license_family":"Custom license","commercial_use":"Yes, with conditions","license_terms":"Companies whose products had more than 700 million monthly active users on the release date need a separate license from Meta. Products must display \"Built with Llama\", model names derived from it must start with \"Llama\", and the Llama Acceptable Use Policy applies.","precision":"BF16","download_gb":217.3,"official_variants":null,"hf_repo":"https://huggingface.co/meta-llama/Llama-4-Scout-17B-16E-Instruct","license_url":"https://github.com/meta-llama/llama-models/blob/main/models/llama4/LICENSE","checked":"2026-09-29","sources":"Hugging Face model page · https://huggingface.co/meta-llama/Llama-4-Scout-17B-16E-Instruct\nHugging Face API · https://huggingface.co/api/models/meta-llama/Llama-4-Scout-17B-16E-Instruct?expand=safetensors"}},{"slug":"command-a","name":"Command A","path":"/open-models/command-a","category":null,"updatedAt":"2026-09-29T18:45:59.488Z","fields":{"developer":"Cohere","released":"March 2025","params":"111B","params_b":111,"active_params":null,"active_b":null,"architecture":"Dense","context":"256K","context_tokens":256000,"modalities":"Text","license":"CC BY-NC 4.0","license_family":"Non-commercial","commercial_use":"No","license_terms":"Non-commercial research use only. The Cohere Labs Acceptable Use Policy also applies.","precision":"BF16","download_gb":222.1,"official_variants":null,"hf_repo":"https://huggingface.co/CohereLabs/c4ai-command-a-03-2025","license_url":"https://creativecommons.org/licenses/by-nc/4.0/","checked":"2026-09-29","sources":"Hugging Face model page · https://huggingface.co/CohereLabs/c4ai-command-a-03-2025\nHugging Face API · https://huggingface.co/api/models/CohereLabs/c4ai-command-a-03-2025?expand=safetensors"}},{"slug":"phi-4","name":"Phi-4","path":"/open-models/phi-4","category":null,"updatedAt":"2026-09-29T18:44:37.626Z","fields":{"developer":"Microsoft","released":"December 12, 2024","params":"14B","params_b":14,"active_params":null,"active_b":null,"architecture":"Dense","context":"16K","context_tokens":16000,"modalities":"Text","license":"MIT","license_family":"MIT","commercial_use":"Yes","license_terms":"MIT: keep the copyright and permission notice.","precision":"BF16","download_gb":29.3,"official_variants":"microsoft/phi-4-gguf; microsoft/phi-4-onnx","hf_repo":"https://huggingface.co/microsoft/phi-4","license_url":"https://huggingface.co/microsoft/phi-4/resolve/main/LICENSE","checked":"2026-09-29","sources":"Hugging Face model card · https://huggingface.co/microsoft/phi-4/raw/main/README.md\nHugging Face API · https://huggingface.co/api/models/microsoft/phi-4?expand=safetensors"}},{"slug":"llama-3-3-70b","name":"Llama 3.3 70B","path":"/open-models/llama-3-3-70b","category":null,"updatedAt":"2026-09-29T18:41:52.587Z","fields":{"developer":"Meta","released":"December 6, 2024","params":"70B","params_b":70,"active_params":null,"active_b":null,"architecture":"Dense","context":"128K","context_tokens":128000,"modalities":"Text","license":"Llama 3.3 Community License","license_family":"Custom license","commercial_use":"Yes, with conditions","license_terms":"Companies whose products had more than 700 million monthly active users on the release date need a separate license from Meta. Products must display \"Built with Llama\", model names derived from it must start with \"Llama\", and the Llama Acceptable Use Policy applies.","precision":"BF16","download_gb":141.1,"official_variants":null,"hf_repo":"https://huggingface.co/meta-llama/Llama-3.3-70B-Instruct","license_url":"https://github.com/meta-llama/llama-models/blob/main/models/llama3_3/LICENSE","checked":"2026-09-29","sources":"Hugging Face model page · https://huggingface.co/meta-llama/Llama-3.3-70B-Instruct\nHugging Face API · https://huggingface.co/api/models/meta-llama/Llama-3.3-70B-Instruct?expand=safetensors"}},{"slug":"llama-3-1-8b","name":"Llama 3.1 8B","path":"/open-models/llama-3-1-8b","category":null,"updatedAt":"2026-09-29T18:41:58.022Z","fields":{"developer":"Meta","released":"July 23, 2024","params":"8B","params_b":8,"active_params":null,"active_b":null,"architecture":"Dense","context":"128K","context_tokens":128000,"modalities":"Text","license":"Llama 3.1 Community License","license_family":"Custom license","commercial_use":"Yes, with conditions","license_terms":"Companies whose products had more than 700 million monthly active users need a separate license from Meta, and the Llama Acceptable Use Policy applies.","precision":"BF16","download_gb":16.1,"official_variants":null,"hf_repo":"https://huggingface.co/meta-llama/Llama-3.1-8B-Instruct","license_url":"https://github.com/meta-llama/llama-models/blob/main/models/llama3_1/LICENSE","checked":"2026-09-29","sources":"Hugging Face model page · https://huggingface.co/meta-llama/Llama-3.1-8B-Instruct\nHugging Face API · https://huggingface.co/api/models/meta-llama/Llama-3.1-8B-Instruct?expand=safetensors"}},{"slug":"llama-3-1-405b","name":"Llama 3.1 405B","path":"/open-models/llama-3-1-405b","category":null,"updatedAt":"2026-09-29T18:42:03.120Z","fields":{"developer":"Meta","released":"July 23, 2024","params":"405B","params_b":405,"active_params":null,"active_b":null,"architecture":"Dense","context":"128K","context_tokens":128000,"modalities":"Text","license":"Llama 3.1 Community License","license_family":"Custom license","commercial_use":"Yes, with conditions","license_terms":"Companies whose products had more than 700 million monthly active users need a separate license from Meta, and the Llama Acceptable Use Policy applies.","precision":"BF16","download_gb":811.7,"official_variants":"meta-llama/Llama-3.1-405B-Instruct-FP8","hf_repo":"https://huggingface.co/meta-llama/Llama-3.1-405B-Instruct","license_url":"https://github.com/meta-llama/llama-models/blob/main/models/llama3_1/LICENSE","checked":"2026-09-29","sources":"Hugging Face model page · https://huggingface.co/meta-llama/Llama-3.1-405B-Instruct\nHugging Face API · https://huggingface.co/api/models/meta-llama/Llama-3.1-405B-Instruct?expand=safetensors\nHugging Face API · https://huggingface.co/api/models?author=meta-llama&search=FP8"}}]}