32 model passports. One consistent evidence model.
Compare architecture, context, licensing, hardware class and deployment routes without flattening provider-specific facts into the base model.
Use the search box to find a model, developer, license family or hardware class. Every row links to a full Model Passport.
| Model | Parameters | Context | License | Hardware class | Passport |
|---|---|---|---|---|---|
| DeepSeek-R1DeepSeek | 671B | 128K tokens | MIT | >80B · datacenter / multi-GPU | Passport → |
| DeepSeek-V3.1DeepSeek | ~671B-class | 128K tokens | MIT | >80B · datacenter / multi-GPU | Passport → |
| Devstral Small 2505Mistral AI | 24B | 128K tokens | Apache 2.0 | 17–32B · high-memory workstation | Passport → |
| Gemma 3 12B ITGoogle DeepMind | 12B | 128K tokens | Gemma Terms | 9–16B · workstation/local | Passport → |
| Gemma 3 27B ITGoogle DeepMind | 27B | 128K tokens | Gemma Terms | 17–32B · high-memory workstation | Passport → |
| Gemma 3 4B ITGoogle DeepMind | 4B | 128K tokens | Gemma Terms | ≤8B · consumer/local | Passport → |
| Gemma 3n E4B ITGoogle DeepMind | ~8B raw / E4B effective profile | 32K tokens | Gemma Terms | ≤8B · consumer/local | Passport → |
| GLM-4.5Z.ai | 355B | 128K tokens | MIT | >80B · datacenter / multi-GPU | Passport → |
| GLM-4.5 AirZ.ai | 106B | 128K tokens | MIT | >80B · datacenter / multi-GPU | Passport → |
| gpt-oss-120bOpenAI | 116.8B | 131,072 tokens | Apache 2.0 | >80B · datacenter / multi-GPU | Passport → |
| gpt-oss-20bOpenAI | 20.9B | 131,072 tokens | Apache 2.0 | 17–32B · high-memory workstation | Passport → |
| Granite 3.3 8B InstructIBM | 8B | 128K tokens | Apache 2.0 | ≤8B · consumer/local | Passport → |
| Kimi K2 InstructMoonshot AI | 1T | 128K tokens | Modified MIT | >80B · datacenter / multi-GPU | Passport → |
| Llama 3.3 70B InstructMeta | 70B | 128K tokens | Llama 3.3 Community | 33–80B · large-memory / multi-GPU | Passport → |
| Llama 4 MaverickMeta | ~400B | 1000000 tokens | Llama 4 Community | >80B · datacenter / multi-GPU | Passport → |
| Llama 4 ScoutMeta | ~109B | 10,000,000 tokens | Llama 4 Community | >80B · datacenter / multi-GPU | Passport → |
| Magistral Small 2506Mistral AI | 24B | 128K nominal · 40K recommended for quality | Apache 2.0 | 17–32B · high-memory workstation | Passport → |
| Mathstral 7B v0.1Mistral AI | 7B | 32,768 tokens | Apache 2.0 | ≤8B · consumer/local | Passport → |
| Mistral Nemo Instruct 2407Mistral AI / NVIDIA | 12B | 128K tokens | Apache 2.0 | 9–16B · workstation/local | Passport → |
| Mistral Small 3.1 24B InstructMistral AI | 24B | 128K tokens | Apache 2.0 | 17–32B · high-memory workstation | Passport → |
| OLMo 2 13B InstructAi2 | 13B | 4,096 tokens | Apache 2.0 | 9–16B · workstation/local | Passport → |
| OLMo 2 32B InstructAi2 | 32B | 4,096 tokens | Apache 2.0 | 17–32B · high-memory workstation | Passport → |
| Phi-4Microsoft | 14B | 16K tokens | MIT | 9–16B · workstation/local | Passport → |
| Phi-4 Mini InstructMicrosoft | 3.8B | 128K tokens | MIT | ≤8B · consumer/local | Passport → |
| Phi-4 Multimodal InstructMicrosoft | 5.6B | 128K tokens | MIT | ≤8B · consumer/local | Passport → |
| Qwen2.5-Coder-32B-InstructQwen / Alibaba | ~32.5B | 32K native · long-context extension documented by Qwen | Apache 2.0 | 17–32B · high-memory workstation | Passport → |
| Qwen2.5-VL-72B-InstructQwen / Alibaba | ~72B | 32K recommended default · config exposes 128K positions | Qwen License | 33–80B · large-memory / multi-GPU | Passport → |
| Qwen3-235B-A22BQwen / Alibaba | 235B | 32,768 native · 131,072 with YaRN | Apache 2.0 | >80B · datacenter / multi-GPU | Passport → |
| Qwen3-30B-A3BQwen / Alibaba | 30.5B | 32,768 native · 131,072 with YaRN | Apache 2.0 | 17–32B · high-memory workstation | Passport → |
| Qwen3-32BQwen / Alibaba | 32.8B | 32,768 native · 131,072 with YaRN | Apache 2.0 | 17–32B · high-memory workstation | Passport → |
| Qwen3-Coder-480B-A35B-InstructQwen / Alibaba | 480B | 256K tokens | Apache 2.0 | >80B · datacenter / multi-GPU | Passport → |
| SmolLM3 3BHugging Face | 3B | 64K native · up to 128K extended | Apache 2.0 | ≤8B · consumer/local | Passport → |
These paths regroup the same verified records around deployment and licensing questions people actually search for.
Compare models by hardware tier, quantization and documented local runtimes.
Open guide →License collectionBrowse the registry models released under Apache License 2.0 and understand the obligations.
Open collection →Commercial useSeparate permissive licenses from custom terms, thresholds and redistribution conditions.
Open guide →It is a structured index of models whose trained weights are obtainable, with separate records for architecture, licensing and deployment evidence.
No. Open weight describes weight availability. Some entries use Apache 2.0 or MIT, while others use custom model terms such as the Llama Community License, Gemma Terms or the Qwen License.
Weight access can make independent deployment possible, but practical feasibility varies from consumer-class 3B models to datacenter-scale mixture-of-experts models.
OpenWeightModels prioritizes exact model cards, license files, official repositories and named deployment-provider documentation, with a visible verification date.