Mistral Small 3.1 (25.03)
Mistral Small 3.1 (25.03) is a text generation model from Mistral with a 128,000-token context window, released in March 2025.
MistralEfficient text generation with long context support
Mistral Small 3.1 (25.03) is a chat-oriented language model developed by Mistral and released in March 2025. It belongs to Mistral's Small model family, designed to balance capability and efficiency within a compact footprint. The model supports a 128,000-token context window, allowing it to process long documents or extended conversations in a single pass, with a maximum response size of 16,000 tokens.
This model is suited for tasks such as text summarization, question answering, instruction following, and general-purpose conversational applications. Its relatively large context window makes it practical for use cases that involve lengthy inputs, such as document analysis or multi-turn dialogue. Mistral provides the model as a first-party offering, meaning it is served directly through Mistral's own infrastructure.
What Mistral Small 3.1 (25.03) supports
Long Context Processing
Handles inputs up to 128,000 tokens in a single request, enabling analysis of long documents or extended conversation histories.
Instruction Following
Responds to structured prompts and multi-step instructions, making it suitable for task-oriented chat applications.
Text Summarization
Condenses long-form content into concise outputs, leveraging the full 128k context window to process entire documents at once.
Code Generation
Generates and explains code across common programming languages as part of its general text generation capabilities.
Multi-turn Dialogue
Maintains coherent conversation across extended exchanges, supported by a 16,000-token maximum response size.
Ready to build with Mistral Small 3.1 (25.03)?
Get Started FreeBenchmark scores
Scores represent accuracy — the percentage of questions answered correctly on each test.
| Benchmark | What it tests | Score |
|---|---|---|
| MMLU-Pro | Expert knowledge across 14 academic disciplines | 52.9% |
| GPQA Diamond | PhD-level science questions (biology, physics, chemistry) | 38.1% |
| MATH-500 | Undergraduate and competition-level math problems | 56.3% |
| AIME 2024 | American math olympiad problems | 6.3% |
| LiveCodeBench | Real-world coding tasks from recent competitions | 14.1% |
| HLE | Questions that challenge frontier models across many domains | 4.3% |
| SciCode | Scientific research coding and numerical methods | 15.6% |
Common questions about Mistral Small 3.1 (25.03)
What is the context window size for Mistral Small 3.1 (25.03)?
Mistral Small 3.1 (25.03) supports a context window of 128,000 tokens, allowing it to process long documents or extended conversations in a single request.
What is the maximum response length this model can produce?
The model has a maximum response size of 16,000 tokens per output.
Who publishes Mistral Small 3.1 (25.03) and how is it served?
The model is published by Mistral and provided as a first-party offering, meaning it is served directly through Mistral's own infrastructure rather than a third-party provider.
When was Mistral Small 3.1 (25.03) released?
Mistral Small 3.1 (25.03) was released in March 2025 and became available on MindStudio on March 21, 2025.
Does Mistral Small 3.1 (25.03) support image or video inputs?
Based on the available metadata, image and video analysis support is not confirmed for this model. It is listed as a text generation model with standard text inputs.
Parameters & options
Explore similar models
MistralMinistral 3 3B
Ministral 3 3B is a 3-billion-parameter open-source text generation model from Mistral with a 256,000-token context window.
MistralMinistral 3 8B
Ministral 3 8B is an open-source text generation model from Mistral with a 256,000-token context window.
MistralMinistral 3 14B
Ministral 3 14B is an open-source text generation model from Mistral with a 256,000-token context window.
OpenAIGPT-6 Astra
AnthropicClaude 5.1 Fable
DeepSeekDeepSeek V4.1 Flash
MistralMistral Nemo
Mistral Nemo is a text generation model from Mistral with a 128,000-token context window, released in July 2024.
Start building with Mistral Small 3.1 (25.03)
No API keys required. Create AI-powered workflows with Mistral Small 3.1 (25.03) in minutes — free.