Mistral Small 3.1

Mistral Small 3.1

Model family: Mistral

Mistral Small 3.1 is Mistral AI’s upgraded Small series model, pairing strong text performance with vision capabilities in a roughly 24-billion parameter package. It can read and discuss images as well as text, supports function calling and structured outputs, and is tuned for tasks like programming, mathematical reasoning, document analysis, and multilingual chat. A long context window allows it to keep track of large conversations or multi document inputs while still decoding quickly enough for interactive use. Because it is released as open weights and optimized for throughput, Mistral Small 3.1 is often chosen as a default engine for production assistants, RAG systems, and code tools where teams want high quality results without the overhead of the very largest models.

Overview

Mistral Small 3.1 is a 24B open weights multimodal model with long context, strong coding and reasoning, and efficient deployment on commodity hardware.

💻Coding 📊Data analysis 💬Chatting 💼Personal assistant

About Mistral AI

Mistral AI is a company that specializes in artificial intelligence and machine learning solutions.

Industry: Artificial Intelligence

Company Size: 316

Location: Paris, FR

Website: mistral.ai

View Company Profile

Other models from this family

Announcement Mistral

Robostral Navigate

By Mistral AI

Robostral Navigate is an 8B parameter robotics navigation model that uses a single RGB camera and natural language instructions to autonomously guide robots through complex environments. It predicts the next target location via image based pointing and falls back to local coordinate displacements when the target is out of view, achieving 76.6 percent success on R2R-CE validation unseen without depth sensors, LiDAR, or multiple cameras.

🤖Autonomous mobile robots 🧩Spatial intelligence 👁️Computer vision assistance 🤖Robotics

NewMultimodal

Released 4d ago
Mistral

Leanstral 1.5 119B A6B

By Mistral AI

Leanstral 1.5 119B A6B is a mixture of experts model with 119 billion total parameters and 6.5 billion active per token, built by Mistral AI as a code agent for formal verification in Lean 4. It accepts text and image input and produces text output, with a 256k token context window. It solves competition math benchmarks like miniF2F and PutnamBench and can verify code correctness properties, uncovering real bugs in open source repositories.

📐Mathematical proofs 🐞Debugging 🤖Agents 🔢Math

NewCoding

Released 10d ago
Mistral

Mistral OCR 4

By Mistral AI

Mistral OCR 4 extracts and structures content from PDF, DOC, PPT, and OpenDocument files, returning text alongside bounding boxes, typed block classification (titles, tables, equations, signatures), and inline confidence scores. Supports 170 languages across 10 language groups. Deployable via API or self-hosted in a single container for data-sovereignty compliance.

🔍Data extraction 📄Document processing 📜OCR

NewMultimodal

Released 19d ago