Mistral Small 3
VerifiedOpen-weight LLM for efficient, versatile text tasks.
About Mistral Small 3
Mistral Small 3 follows a transformer-based design focused on text inputs and outputs. Its open-weight release allows full access to model parameters for inspection and modification. The 32768-token context enables handling of longer documents without truncation.
Strengths include deployment flexibility on varied hardware and support for fine-tuning on domain-specific data. Typical usage covers chat interfaces, summarization pipelines, and code assistance tools. Users integrate it into applications where transparency and local control matter most.
Capabilities
Benchmarks & performance
Independent evaluation scores and measured speed.
Source: Artificial Analysis
How Mistral Small 3 compares
Mistral Small 3 (striped bar) vs other language models on intelligence, speed and price.
Intelligence
Artificial Analysis Intelligence Index · Higher is better · Mistral Small 3 ranks #56 of 67
Speed
Output tokens per second · Higher is better · Mistral Small 3 ranks #9 of 45
Price
USD per 1M output tokens · Lower is better · Mistral Small 3 ranks #5 of 141
Sources: Artificial Analysis (intelligence, speed) · OpenRouter (price).
Best for
Code Generation and Completion
Mistral Small 3 performs well on code generation and completion tasks, producing functional snippets and completing partial code in multiple programming languages.
Multilingual Text Processing
The model handles multilingual text processing effectively, supporting generation, translation, and analysis across various languages in a single workflow.
Document Summarization and Reasoning
It delivers strong results in text summarization and logical reasoning, condensing long inputs while maintaining key arguments and conclusions.
Strengths & limitations
Strengths
- +Efficient and fast inference
- +Cost-effective for general use
- +Solid performance relative to model size
- +Reliable on everyday language tasks
Limitations
- –Text-only modality
- –Less capable on highly complex or specialized reasoning
- –32k context limits very long-document handling
Cost calculator
Estimate what Mistral Small 3 would cost for your usage.
Based on Mistral Small 3's $0.05/1M input · $0.08/1M output. Estimate only — actual cost varies by provider and caching.
Quick start
OpenRouter's API is OpenAI-compatible — most SDKs work by just swapping the base URL. Only the model slug changes between models.
import OpenAI from "openai";
const client = new OpenAI({
baseURL: "https://openrouter.ai/api/v1",
apiKey: process.env.OPENROUTER_API_KEY,
});
const completion = await client.chat.completions.create({
model: "mistralai/mistral-small-24b-instruct-2501",
messages: [{ role: "user", content: "Hello!" }],
});
console.log(completion.choices[0].message.content);Model slug: mistralai/mistral-small-24b-instruct-2501
Editor's verdict
Mistral Small 3 is Mistral's open-weight language models with a 33K-token context window.
On independent testing it scores 12.7 on the Artificial Analysis Intelligence Index, running at roughly 163 tokens per second with about 0.74s to first token.
At $0.08 per 1M output tokens, it is very cost-efficient for its class.
As an open-weight model you can self-host it or call it through a hosted API.
Best suited to efficient and fast inference and cost-effective for general use.
Frequently asked questions
The model provides a context window of 32768 tokens.
User reviews
Real, verified reviews from the community shape this model's rating.
Loading reviews…
Other Mistral models
Sibling versions in the Mistral family from Mistral.
Similar models
Other language models worth comparing.