Model results
GPQA Diamond
Choose a shared evaluation protocol to compare model scores. Versions, sources and configurations stay separate.
Atlas snapshot captured 10/19/2026, 10:20:28 AM UTC. This is an import timestamp, not the evaluation date. Coverage may be incomplete.
Reviewed recommendation and limits
Use as a compact reference for expert-written science questions, alongside evidence that matches the scientific work you intend to do.
Original GPQA Diamond subset. Preserve prompt, answer extraction, tools, sampling and aggregation. Keep main, extended and third-party corrected subsets separate.. Imported groups below may use different protocols; this review does not certify each run. The static 198-question subset has limited statistical power and exposure risk. Its expert filtering does not make it representative of scientific practice. A Google-proof collection process does not prove present-day contamination resistance.
GPQA Diamond
- Source
- Epoch AI Benchmarking Hub
- Version
- Not reported
- Measurement date
- Not reported
- Metric
- % · Higher is better
- Evaluation scope
- Epoch AI run · task v1 · effort minimal · result date not supplied
Exact score ties share a position. Positions do not establish statistical significance. Missing metadata is shown as unreported; matching imported metadata is not independent verification of the protocol.
#1Gemini 3.5 Flash
google/gemini-3.5-flash-20260519
86.4%
View modelEvidence independent#2Gemini 3.6 Flash
google/gemini-3.6-flash-20260721
85.9%
View modelEvidence independent#3Gemma 4 31B
google/gemma-4-31b-it-20260402
75.8%
View modelEvidence independent#4Gemini 3.5 Flash Lite
google/gemini-3.5-flash-lite-20260721
74.2%
View modelEvidence independent#5Gemini 3.1 Flash Lite
google/gemini-3.1-flash-lite-20260507
73.7%
View modelEvidence independent#6Gemma 4 26B A4B
google/gemma-4-26b-a4b-it-20260403
73.2%
View modelEvidence independent#7GPT-5
openai/gpt-5-2025-08-07
71.7%
View modelEvidence independent#7GPT-5 Mini
openai/gpt-5-mini-2025-08-07
71.7%
View modelEvidence independent#9GPT-5 Nano
openai/gpt-5-nano-2025-08-07
48.5%
View modelEvidence independent