in the loop

Mistral Medium 3.5
#110 of 274 scored.

MistralApr 28, 2026Open weights

Capabilities Index

Range 138.6–143.2

141.3

Context

Takes text, image, file

262K

Input

Per 1M tokens

$1.50

Output

Per 1M tokens

$7.50

Benchmarks

No independent scores yet.

Epoch AI hasn't run Mistral Medium 3.5 yet. Its scores show up here the morning after they do.

Everything else Epoch has for it

  • DTBench59.1%
  • WeirdML43.7%
  • LMCA30.7%
  • Surface Evolver Bench26.9%
  • ProofBench9.0%
  • FrontierCode8.0%

On the index

Around it.

  1. #107Gemma 4 26B A4BGoogle · $0.068 in141.8
  2. #108Claude Sonnet 4Anthropic · $3.00 in141.7
  3. #109Gemini 2.5 Flash (May 2025)Google141.5
  4. #110Mistral Medium 3.5Mistral · $1.50 in141.3
  5. #111DeepSeek-R1 (May 2025)DeepSeek141.3
  6. #112Claude 3.7 SonnetAnthropic141.2
  7. #113Gemini 2.5 Flash (Jun 2025)Google140.8

Sources