Mistral Models
At Mistral, we believe in the power of openness and broad distribution to promote innovation and collaboration in AI. Based on this principle, we release models to the community that are open for anyone to use, subject to the terms of the license. These models supports wide range of use cases:
- Text generation and chat
- Code generation and editing
- Math reasoning
- Multimodal generation and chat
- Coding agent.
To learn more about all of our models, please visit our website.
List of Open models
| Model | Task | Text | Image | Tokenizer | Context size | License | HF Hub Repo |
|---|---|---|---|---|---|---|---|
| Mistral 7B v0.1 | Generation | ✅ | ❌ | v1 | 8k | Apache 2.0 | mistralai/Mistral-7B-v0.1 |
| Mistral 7B Instruct v0.1 | Instruct | ✅ | ❌ | v1 | 8k | Apache 2.0 | mistralai/Mistral-7B-Instruct-v0.1 |
| Mixtral 8x7B | Generation | ✅ | ❌ | v1 | 32k | Apache 2.0 | mistralai/Mixtral-8x7B-v0.1 |
| Mixtral 8x7B Instruct v0.1 | Instruct | ✅ | ❌ | v1 | 32k | Apache 2.0 | mistralai/Mixtral-8x7B-Instruct-v0.1 |
| Mistral 7B Instruct v0.2 | Instruct | ✅ | ❌ | v1 | 32k | Apache 2.0 | mistralai/Mistral-7B-Instruct-v0.2 |
| Mixtral 8x22B v0.1 | Generation | ✅ | ❌ | v1 | 65k | Apache 2.0 | mistralai/Mixtral-8x22B-v0.1 |
| Mixtral 8x22B Instruct v0.1 | Instruct | ✅ | ❌ | v1 | 65k | Apache 2.0 | mistralai/Mixtral-8x22B-Instruct-v0.1 |
| Mistral 7B v0.3 | Generation | ✅ | ❌ | v3 | 32k | Apache 2.0 | mistralai/Mistral-7B-v0.3 |
| Mistral 7B v0.3 Instruct | Instruct | ✅ | ❌ | v3 | 32k | Apache 2.0 | mistralai/Mistral-7B-Instruct-v0.3 |
| Codestral 22B v0.1 | Instruct and FIM | ✅ | ❌ | v3 | 32k | MNPL | mistralai/Codestral-22B-v0.1 |
| Mathstral 7B v0.1 | Instruct | ✅ | ❌ | v3 | 32k | Apache 2.0 | mistralai/Mathstral-7B-v0.1 |
| Mamba Codestral 7B v0.1 | Instruct and FIM | ✅ | ❌ | v3 | 256k | Apache 2.0 | mistralai/Mamba-Codestral-7B-v0.1 |
| Mistral Nemo Base 2407 | Generation | ✅ | ❌ | v3 - tekken | 128k | Apache 2.0 | mistralai/Mistral-Nemo-Base-2407 |
| Mistral Nemo Instruct 2407 | Instruct | ✅ | ❌ | v3 - tekken | 128k | Apache 2.0 | mistralai/Mistral-Nemo-Instruct-2407 |
| Mistral Large Instruct 2407 | Instruct | ✅ | ❌ | v3 - tekken | 128k | MRL | mistralai/Mistral-Large-Instruct-2407 |
| Pixtral 12B Base 2409 | Generation | ✅ | ✅ | v3 - tekken | 128k | Apache 2.0 | mistralai/Pixtral-12B-Base-2409 |
| Pixtral 12B 2409 | Instruct | ✅ | ✅ | v3 - tekken | 128k | Apache 2.0 | mistralai/Pixtral-12B-2409 |
| Mistral Large Instruct 2411 | Instruct | ✅ | ❌ | v7 | 128k | MRL | mistralai/Mistral-Large-Instruct-2411 |
| Pixtral Large Instruct 2411 | Instruct | ✅ | ✅ | v7 | 128k | MRL | mistralai/Pixtral-Large-Instruct-2411 |
| Mistral Small 24B Base 2501 | Generation | ✅ | ❌ | v7-tekken | 32k | Apache 2.0 | mistralai/Mistral-Small-24B-Base-2501 |
| Mistral Small 24B Instruct 2501 | Instruct | ✅ | ❌ | v7-tekken | 32k | Apache 2.0 | mistralai/Mistral-Small-24B-Instruct-2501 |
| Mistral Small 3.1 24B Base 2503 | Generation | ✅ | ✅ | v7-tekken | 128k | Apache 2.0 | mistralai/Mistral-Small-3.1-24B-Base-2503 |
| Mistral Small 3.1 24B Instruct 2503 | Instruct | ✅ | ✅ | v7-tekken | 128k | Apache 2.0 | mistralai/Mistral-Small-3.1-24B-Instruct-2503 |
| Devstral Small 2505 | Coding Agent | ✅ | ❌ | v7-tekken | 128k | Apache 2.0 | mistralai/Devstral-Small-2505 |