Skip to content
Slow Variables

Ask the data

Answers come from the same store as the site; every number is checked against the record it cites.

Indicators / series

epoch_price.humaneval_gpt_4_0314.lowest_usd_per_mtok.pt

Source: Epoch AI LLM inference price trends (Epoch AI) · CC BY 4.0 · CSV

Epoch AI, LLM inference price trends (CC BY 4.0)

idvaluegradeas of datepublished datetieraudited vs reportedextraction methodreview statusretrieved athttp statuscontent hashsupersedes idurlflags
1793df9a7fe54e16$37.5grade B, evidence tier 62023-03-142023-03-146reportedapiapproved2026-09-10T06:33:11.138943Z200cb4f44d88b39source page
bdaf0970dff30d54$3.25grade B, evidence tier 62023-06-132023-06-136reportedapiapproved2026-09-10T06:33:11.138943Z200cb4f44d88b39source page
cab59f833c24feb5$0.75grade B, evidence tier 62023-11-062023-11-066reportedapiapproved2026-09-10T06:33:11.138943Z200cb4f44d88b39source page
31b02a26808d7389$0.50grade B, evidence tier 62024-03-042024-03-046reportedapiapproved2026-09-10T06:33:11.138943Z200cb4f44d88b39source page
443b06cdfd8e8426$0.26grade B, evidence tier 62024-06-242024-06-246reportedapiapproved2026-09-10T06:33:11.138943Z200cb4f44d88b39source page
6ca37ca91a6bde2c$0.13grade B, evidence tier 62024-07-182024-07-186reportedapiapproved2026-09-10T06:33:11.138943Z200cb4f44d88b39source page
9bac57ffb70310e8$0.10grade B, evidence tier 62024-07-232024-07-236reportedapiapproved2026-09-10T06:33:11.138943Z200cb4f44d88b39source page
Raw snippets and dispute text
  • 1793df9a7fe54e16 {"Benchmark": "HumanEval", "Benchmark score": "67.0", "Model Name": "GPT-4-0314", "Release Date": "2023-03-14", "Threshold model": "GPT-4-0314", "USD per 1M Tokens": "37.5"}
  • bdaf0970dff30d54 {"Benchmark": "HumanEval", "Benchmark score": "71.0", "Model Name": "GPT-3.5-Turbo-2023-06", "Release Date": "2023-06-13", "Threshold model": "GPT-4-0314", "USD per 1M Tokens": "3.25"}
  • cab59f833c24feb5 {"Benchmark": "HumanEval", "Benchmark score": "71.0", "Model Name": "GPT-3.5-Turbo-2023-11", "Release Date": "2023-11-06", "Threshold model": "GPT-4-0314", "USD per 1M Tokens": "0.75"}
  • 31b02a26808d7389 {"Benchmark": "HumanEval", "Benchmark score": "77.0", "Model Name": "Claude-3-Haiku", "Release Date": "2024-03-04", "Threshold model": "GPT-4-0314", "USD per 1M Tokens": "0.5"}
  • 443b06cdfd8e8426 {"Benchmark": "HumanEval", "Benchmark score": "76.0", "Model Name": "Gemma-2-27B", "Release Date": "2024-06-24", "Threshold model": "GPT-4-0314", "USD per 1M Tokens": "0.26"}
  • 6ca37ca91a6bde2c {"Benchmark": "HumanEval", "Benchmark score": "71.0", "Model Name": "Mistral-NeMo", "Release Date": "2024-07-18", "Threshold model": "GPT-4-0314", "USD per 1M Tokens": "0.13"}
  • 9bac57ffb70310e8 {"Benchmark": "HumanEval", "Benchmark score": "67.0", "Model Name": "Llama-3.1-Instruct-8B", "Release Date": "2024-07-23", "Threshold model": "GPT-4-0314", "USD per 1M Tokens": "0.1"}