in the loop

Mistral Large 2 (Jul 2024)
#170 of 274 scored.

MistralJul 24, 2024Open weights

Capabilities Index

Range 122.0–129.6

127.5

Context

Tokens

—

Input

Per 1M tokens

—

Output

Per 1M tokens

—

Benchmarks

How it scores.

  • GPQA Diamond32.0%

    137th of 186

  • Mock AIME8.4%

    136th of 176

Everything else Epoch has for it

  • MMLU73.3%
  • Lech Mazur Writing69.0%
  • MATH level 544.8%
  • DTBench35.4%
  • SimpleBench7.0%

On the index

Around it.

  1. #167Llama 3.1-405BMeta128.8
  2. #168Mistral Large 2 (Nov 2024)Mistral128.5
  3. #169Qwen2.5-32BAlibaba128.5
  4. #170Mistral Large 2 (Jul 2024)Mistral127.5
  5. #171Mistral Small 3.1Mistral127.5
  6. #172Llama 3.3 70BMeta127.3
  7. #173GPT-4 Turbo (Apr 2024)OpenAI127.3

Sources