Mistral Medium 3.5 Official Specifications, Pricing & Benchmarks
Mistral Medium 3.5 (mistral-medium-2604) by Mistral AI, released 2026-04-29, featuring 262,144 context window, 262,144 max output tokens, and Text+Image+PDF input.
📋 Mistral Medium 3.5 Official API & System Card Specifications
Verified against official Mistral AI developer documentation & pricing
1. Identity & Capabilities
API Spec- API Model ID
- mistral-medium-2604
- Release Date
- 2026-04-29
- Weight License
- Open-Weight (API Spec)
- Native Reasoning
- ✅ Supported
- Function Calling & JSON Schema
- ✅ 原生支持
- Batch API 异步批处理
- ✅ Supported (50% off)
2. Context, Output & Modalities
Limits- Max Context Window
- 262,000 tokens (262K)
- Max Output Tokens
- 262,144 tokens
- Input Modalities
- Text + Image + PDF
- Output Modalities
- Text
- Direct China Access
- 🌐 Global / OpenRouter
- Deployment Mode
- Open-Weight + Cloud API
3. Official Pricing & Cache
USD / 1M- Input / 1M Tokens
- $ 1.5
- Cached Input / 1M
- $ 0.15 -90%
- Output / 1M Tokens
- $ 7.5
🎛️ Official Sampling & Reasoning Parameters
Recommended generation settings & API parameter constraints
📊 Mistral Medium 3.5 Full Benchmark Scorecard
8 public benchmarks · ranks among 154 models · synced 2026-11-20
🤖 Agents & Tools 3 items
BrowseComp is a benchmark comprising 1,266 questions that challenge AI agents to persistently navigate the internet in search of hard-to-find, entangled information.
Finance Agent v2 is an agentic financial-analysis benchmark from Vals that evaluates models on real-world finance workflows, measuring their ability to retrieve and reason over financial documents, perform multi-step calculations, and produce accurate analyses.
τ³-Bench banking domain evaluates agentic models on multi-turn, tool-using customer-support scenarios in a simulated retail banking environment.
📐 Math 2 items
All 30 problems from the 2025 American Invitational Mathematics Examination (AIME I and AIME II), testing olympiad-level mathematical reasoning with integer answers from 000-999.
Beyond AIME is a difficult mathematical reasoning benchmark designed to test deeper reasoning chains and harder decomposition than standard AIME-style problem sets.
🧠 Reasoning & Science 1 items
📚 Knowledge & General 1 items
Source: ZeroEval / llm-stats.com (✓ = third-party verified, others vendor-reported).
Closest Self-Hostable Open-Weight Alternatives to Mistral Medium 3.5
Open-weight models with full HuggingFace weights for private GPU / Mac deployment
Frequently Asked Questions
Is Mistral Medium 3.5 open-source or proprietary?
Mistral Medium 3.5 (mistral-medium-2604) is a proprietary commercial model by Mistral AI. Weights and internal layer dimensions are undisclosed; it is accessed via cloud API.
What are the context limit and API pricing of Mistral Medium 3.5?
Mistral Medium 3.5 supports 262K input context and 262,144 max output tokens, priced at $1.5/1M input ($0.15/1M cached) and $7.5/1M output.