Mistral Small 4
VerifiedOpen-weight multimodal model for long-context text and image tasks.
About Mistral Small 4
Mistral Small 4 uses a multimodal architecture that integrates text and vision capabilities. Its 262144-token context window enables processing of lengthy documents paired with images. The model is released as open weights, allowing broad access and customization.
Its primary strengths lie in handling extended multimodal sequences without truncation. This design supports coherent reasoning across large amounts of combined textual and visual data. Users benefit from its open-weight availability for local deployment and fine-tuning.
Typical usage includes analyzing long reports with embedded visuals, processing image-rich conversations, and building applications that require sustained context across modalities. Developers often integrate it into systems needing both vision and language understanding over extended inputs.
Capabilities
How Mistral Small 4 compares
Mistral Small 4 (striped bar) vs other multimodal on intelligence, speed and price.
Price
USD per 1M output tokens · Lower is better · Mistral Small 4 ranks #10 of 63
Sources: Artificial Analysis (intelligence, speed) · OpenRouter (price).
Best for
Long-document image analysis
The model processes images embedded within extremely long texts, delivering detailed descriptions and insights while maintaining coherence across 262144 tokens.
Extended multimodal reasoning
It handles complex cross-modal instructions that combine vision and language inputs over large contexts, supporting tasks like summarizing illustrated reports or technical manuals.
Vision-language content generation
Users can provide images and lengthy textual prompts to generate extended, contextually accurate responses that integrate visual understanding with detailed text output.
Strengths & limitations
Strengths
- +Very large 256k token context window
- +Native text and image support
- +Efficient multimodal architecture
Limitations
- –Small model size may limit depth on complex tasks
- –Supports only text and image modalities
- –No audio or video capabilities
Pricing by provider
Live per-provider pricing & uptime, routed via OpenRouter. Prices are USD per 1M tokens.
| Provider | Input /1M | Output /1M | Context | Uptime |
|---|---|---|---|---|
| Mistral | $0.15 | $0.60 | 262K | 100.0% |
| Venice(fp8) | $0.19 | $0.75 | 256K | 100.0% |
Cost calculator
Estimate what Mistral Small 4 would cost for your usage.
Based on Mistral Small 4's $0.15/1M input · $0.60/1M 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-2603",
messages: [{ role: "user", content: "Hello!" }],
});
console.log(completion.choices[0].message.content);Model slug: mistralai/mistral-small-2603
Editor's verdict
Mistral Small 4 is Mistral's open-weight multimodal with a 262K-token context window.
At $0.60 per 1M output tokens, it is very cost-efficient for its class, served by 2 providers.
As an open-weight model you can self-host it or call it through a hosted API.
Best suited to very large 256k token context window and native text and image support.
Frequently asked questions
The model supports a context window of 262144 tokens for handling extended inputs and outputs.
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 multimodal worth comparing.