Mistral Small 3.2 24B
VerifiedOpen-weight multimodal model for efficient image and text tasks.
About Mistral Small 3.2 24B
Mistral Small 3.2 24B combines vision and language processing in a single 24B parameter architecture. It is released as open weights to allow full inspection, fine-tuning, and local deployment. The design supports both image and text modalities through a shared context window of 128000 tokens.
Its strengths center on flexibility for multimodal workloads without requiring proprietary APIs. The model balances parameter count and capability to run on accessible hardware while retaining strong performance across vision-language tasks. Open availability encourages community-driven adaptations and research.
Typical usage includes visual question answering, document understanding, and image captioning pipelines. Developers integrate it into applications needing simultaneous analysis of text and images. Fine-tuning on domain-specific datasets is a common workflow for production deployments.
Capabilities
How Mistral Small 3.2 24B compares
Mistral Small 3.2 24B (striped bar) vs other multimodal on intelligence, speed and price.
Price
USD per 1M output tokens · Lower is better · Mistral Small 3.2 24B ranks #8 of 124
Sources: Artificial Analysis (intelligence, speed) · OpenRouter (price).
Best for
Long Multimodal Document Analysis
Processes extensive reports or articles containing both text and embedded images within its 128,000-token context for summarization and extraction tasks.
Extended Visual Reasoning Conversations
Maintains coherence across lengthy dialogues that combine image inputs with detailed textual follow-up questions and instructions.
Multimodal Content Creation Workflows
Supports iterative generation and refinement of text paired with visual elements over long sessions without losing earlier context.
Strengths & limitations
Strengths
- +Efficient inference suitable for deployment
- +Solid integration of vision and language
- +Supports extended 128k context windows
- +Good balance of capability and speed
Limitations
- –Smaller model size limits performance on hardest tasks
- –Restricted to image and text modalities only
- –May underperform larger models in nuanced reasoning
Cost calculator
Estimate what Mistral Small 3.2 24B would cost for your usage.
Based on Mistral Small 3.2 24B's $0.07/1M input · $0.20/1M 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-3.2-24b-instruct",
messages: [{ role: "user", content: "Hello!" }],
});
console.log(completion.choices[0].message.content);Model slug: mistralai/mistral-small-3.2-24b-instruct
Editor's verdict
Mistral Small 3.2 24B is Mistral's open-weight multimodal with a 128K-token context window.
At $0.20 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 inference suitable for deployment and solid integration of vision and language.
Frequently asked questions
The model supports a context window of 128,000 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 multimodal worth comparing.