in the loop

GPT-4 (Jun 2023)
#187 of 274 scored.

OpenAIJun 13, 2023Closed

Capabilities Index

Range 116.6–126.5

123.1

Context

Tokens

—

Input

Per 1M tokens

—

Output

Per 1M tokens

—

Benchmarks

How it scores.

  • GPQA Diamond7.5%

    171st of 186

  • Mock AIME1.0%

    169th of 176

Everything else Epoch has for it

  • GSM8K90.0%
  • TriviaQA84.8%
  • MMLU76.5%
  • BBH66.8%
  • DTBench37.8%
  • METR Time Horizons29.3%
  • MATH level 523.0%
  • LMCA20.1%
  • WeirdML12.4%
  • Mystery Game Puzzles3.1%
  • Chess Puzzles0.0%

On the index

Around it.

  1. #184DeepSeek-V2 (MoE-236B, May 2024)DeepSeek124.8
  2. #185Amazon Nova ProAmazon123.8
  3. #186Gemma 3 12BGoogle · $0.05 in123.5
  4. #187GPT-4 (Jun 2023)OpenAI123.1
  5. #188Llama 3-70BMeta122.9
  6. #189Gemini 1.5 Flash (May 2024)Google122.6
  7. #190Gemma 2 27BGoogle · $0.65 in122.1

Sources