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.math_5_gpt_4_0613.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
2bc7c25e28d853cf$37.5grade B, evidence tier 62023-06-132023-06-136reportedapiapproved2026-09-10T06:33:11.138943Z200cb4f44d88b39source page
372d933f1d95fcbd$15.0grade B, evidence tier 62023-11-062023-11-066reportedapiapproved2026-09-10T06:33:11.138943Z200cb4f44d88b39source page
156f537e047eb4bd$1.20grade B, evidence tier 62024-04-172024-04-176reportedapiapproved2026-09-10T06:33:11.138943Z200cb4f44d88b39source page
59e63fc3a2b22246$0.13grade B, evidence tier 62024-05-102024-05-106reportedapiapproved2026-09-10T06:33:11.138943Z200cb4f44d88b39source page
0e8c73c17c1caee3$0.12grade B, evidence tier 62024-12-132024-12-136reportedapiapproved2026-09-10T06:33:11.138943Z200cb4f44d88b39source page
Raw snippets and dispute text
  • 2bc7c25e28d853cf {"Benchmark": "MATH 5", "Benchmark score": "23.0", "Model Name": "GPT-4-0613", "Release Date": "2023-06-13", "Threshold model": "GPT-4-0613", "USD per 1M Tokens": "37.5"}
  • 372d933f1d95fcbd {"Benchmark": "MATH 5", "Benchmark score": "36.0", "Model Name": "GPT-4 Turbo", "Release Date": "2023-11-06", "Threshold model": "GPT-4-0613", "USD per 1M Tokens": "15.0"}
  • 156f537e047eb4bd {"Benchmark": "MATH 5", "Benchmark score": "23.0", "Model Name": "Mistral-8x22", "Release Date": "2024-04-17", "Threshold model": "GPT-4-0613", "USD per 1M Tokens": "1.2"}
  • 59e63fc3a2b22246 {"Benchmark": "MATH 5", "Benchmark score": "23.0", "Model Name": "Gemini-1.5-Flash-2024-05", "Release Date": "2024-05-10", "Threshold model": "GPT-4-0613", "USD per 1M Tokens": "0.13"}
  • 0e8c73c17c1caee3 {"Benchmark": "MATH 5", "Benchmark score": "65.0", "Model Name": "Phi 4", "Release Date": "2024-12-13", "Threshold model": "GPT-4-0613", "USD per 1M Tokens": "0.1225"}