modelgrep

Longest-Context Mistral Models

Quick answer · Updated June 2026

Mistral Medium 3.5 has the largest context window of any Mistral model, at 262K tokens. Mistral Small 4 (262K) and Devstral 2 2512 (262K) round out the top three.

262KContext
39.2Intelligence
$1.50Input /M

AI models with the largest context windows, ranked by token capacity. The best large language models for long documents, codebases and extended conversations.

  1. 1M
    mistral-medium-3-5
    ReasoningToolsJSON+139.2 intel · $1.50/M
    262K
    Context
  2. 2M
    mistral-small-2603
    ReasoningToolsJSON+118.6 intel · $0.150/M · 110 t/s
    262K
    Context
  3. 3M
    devstral-2512
    ToolsJSON22.0 intel · $0.400/M · 10 t/s
    262K
    Context
  4. 4M
    ministral-14b-2512
    ToolsJSONVision16.0 intel · $0.200/M · 50 t/s
    262K
    Context
  5. 5M
    ministral-8b-2512
    ToolsJSONVision14.8 intel · $0.150/M · 11 t/s
    262K
    Context
  6. 6M
    mistral-large-2512
    ToolsJSONVision22.8 intel · $0.500/M · 37 t/s
    262K
    Context
  7. 7M
    codestral-2508
    ToolsJSON$0.300/M · 67 t/s · 172ms ttft
    256K
    Context

Frequently asked

Which Mistral model has the largest context window?

Mistral Medium 3.5 has the largest context window of any Mistral model, at 262K tokens. Mistral Small 4 (262K) and Devstral 2 2512 (262K) round out the top three.

What's a good alternative to Mistral Medium 3.5?

Mistral Small 4 (262K) is the closest alternative on this metric, followed by Devstral 2 2512 (262K). See the full ranking above for the tradeoffs.

How many Mistral models are there?

modelgrep tracks 19 Mistral models with live benchmarks, speed, latency and per-provider pricing, led on intelligence by Mistral Medium 3.5. 7 of them qualify for this ranking.

More Mistral rankings

All rankings