Leaderboard
Humanity's Last Examofficial site ↗
22 models scored · metric: accuracy. Prices are the cheapest listed across all serving providers. Cost/run and Pts/$ are priced for the heavy reasoning token profile — thinking tokens outnumber the answer several times over.
A benchmark is priced for the workload it exercises, not a fixed input:output blend: an agent loop and a single hard question bill very differently on the same rate card. Profile: 2k in · 1.5k out · 8k reasoning · 0% cache hit · 12k context.
Scores are facts as publicly reported by labs (see report links below and on each model page) and aggregated via models.dev. Humanity's Last Exam is maintained by its own project — Model Pulse is not affiliated with or endorsed by it.
- 1.Nemotron 3 Ultra 550B A55B32,522
- 2.GLM-5.25,173
- 3.GPT-5.54,719
- 4.Step 3.7 Flash4,324
- 5.Gemini 3.5 Flash3,669
Score reports: anthropic.com · anthropic.com · ai.meta.com · openai.com · openai.com · alibabacloud.com — full source URLs are linked on each model page.
Cheapest scorer: Nemotron 3 Ultra 550B A55B at $0.1 input /M.