in the loop

Mistral Small 3
#175 of 274 scored.

MistralJan 30, 2025Open weights

Capabilities Index

Range 121.4–128.8

127.1

Context

Takes text

33K

Input

Per 1M tokens

$0.05

Output

Per 1M tokens

$0.08

Benchmarks

How it scores.

  • GPQA Diamond29.7%

    146th of 186

  • Mock AIME6.6%

    145th of 176

Everything else Epoch has for it

  • MATH level 544.8%
  • Chess Puzzles0.0%

On the index

Around it.

  1. #172Llama 3.3 70BMeta127.3
  2. #173GPT-4 Turbo (Apr 2024)OpenAI127.3
  3. #174Claude 3.5 HaikuAnthropic127.2
  4. #175Mistral Small 3Mistral · $0.05 in127.1
  5. #176Gemini 1.5 Pro (May 2024)Google126.9
  6. #177Claude 3 OpusAnthropic126.9
  7. #178GPT-4o miniOpenAI · $0.15 in126.6

Sources