Plain-English Summary
A compact 8B Mistral model designed for local, edge, and private multimodal applications.
Ministral 3 8B
A compact 8B Mistral model designed for local, edge, and private multimodal applications.
Ministral 3
Multimodal LLM
GA / open weights
Open weights: Yes
API: Yes and self-hosted
Technical Notes
8B multimodal transformer with 256K context, text-and-image input, tool use, and efficient local deployment.
Best For
On-device or private assistants, edge RAG, extraction, multimodal apps
Watch Out For
Lower reasoning and knowledge capacity than larger Mistral models.
Model Specs
- Maker
- Mistral AI
- Country
- France
- Family
- Ministral 3
- Type
- Multimodal LLM
- Release date
- 2025-12-02
- Date confidence
- Exact
- Status
- GA / open weights
- Context window
- 262,144 tokens
- Max output
- Not listed
- Parameters
- 8B
- Active parameters
- 8B
- Architecture
- Dense multimodal transformer
- Reasoning
- General reasoning
- Tool calling
- Yes
- Structured output
- Yes
- API available
- Yes and self-hosted
- Open weights
- Yes
- Self-hostable
- Yes
- License
- Apache 2.0
- Inputs
- Text, image
- Outputs
- Text
- Approx. price
- $0.20 input / $0.20 output per 1M tokens (14B listing; check tier)
- How to access
- Provider website
- Model / API ID
- ministral-3-8b-25-12
- Knowledge cutoff
- Not publicly disclosed
- Fine-tuning
- Fine-tuning, quantization, local deployment
- Completeness
- 95 %
- Verified
- Jul 18, 2026
- Snapshot date
- Jul 18, 2026
- Source
- Official source