Plain-English Summary
A compact hybrid model that combines general chat, coding, agents, and reasoning in one deployable package.
Mistral Small 4
A compact hybrid model that combines general chat, coding, agents, and reasoning in one deployable package.
Mistral Small
Multimodal LLM
GA / open weights
Open weights: Yes
API: Yes and self-hosted
Technical Notes
Open multimodal model with 256K context and multiple operating modes for standard, coding, agentic, and reasoning tasks.
Best For
Private deployment, coding assistants, agents, multimodal enterprise apps
Watch Out For
Smaller capability ceiling than Medium or Large models.
Model Specs
- Maker
- Mistral AI
- Country
- France
- Family
- Mistral Small
- Type
- Multimodal LLM
- Release date
- 2026-03-16
- Date confidence
- Exact
- Status
- GA / open weights
- Context window
- 262,144 tokens
- Max output
- Not listed
- Parameters
- Not disclosed
- Active parameters
- Not disclosed
- Architecture
- Hybrid multimodal transformer
- Reasoning
- Yes — selectable modes
- Tool calling
- Yes
- Structured output
- Yes
- API available
- Yes and self-hosted
- Open weights
- Yes
- Self-hostable
- Yes
- License
- Open-weight license
- Inputs
- Text, image
- Outputs
- Text
- Approx. price
- See provider pricing
- How to access
- Provider website
- Model / API ID
- mistral-small-4
- Knowledge cutoff
- Not publicly disclosed
- Fine-tuning
- Fine-tuning, quantization, self-hosting
- Completeness
- 86 %
- Verified
- Jul 18, 2026
- Snapshot date
- Jul 18, 2026
- Source
- Official source