in the loop

GPT-4 Turbo (Nov 2023)
#179 of 274 scored.

OpenAIJan 25, 2024Closed

Capabilities Index

Range 120.9–129.0

126.5

Context

Tokens

—

Input

Per 1M tokens

—

Output

Per 1M tokens

—

Benchmarks

How it scores.

  • GPQA Diamond23.2%

    153rd of 186

Everything else Epoch has for it

  • MMLU72.8%
  • MATH level 540.0%
  • DTBench36.0%
  • METR Time Horizons35.2%
  • LMCA11.5%

On the index

Around it.

  1. #176Gemini 1.5 Pro (May 2024)Google126.9
  2. #177Claude 3 OpusAnthropic126.9
  3. #178GPT-4o miniOpenAI · $0.15 in126.6
  4. #179GPT-4 Turbo (Nov 2023)OpenAI126.5
  5. #180Llama 3.1-70BMeta125.9
  6. #181GPT-4 (Mar 2023)OpenAI125.9
  7. #182Llama 3.2 90BMeta125.5

Sources