Codesota · Benchmark · GPQAHome/Leaderboards/GPQA
Unknown

GPQA.

448 expert-level questions in biology, physics, and chemistry. Designed to be unsearchable.

Paper ↗Leaderboard ↓Lineage
§ 01 · Leaderboard

Results by metric.

Found a wrong score or missing run?
Use row edits to send a sourced correction into moderation.
Add / edit result ↗Report issue ↗

accuracy

Accuracy is the reported evaluation metric for GPQA. Codesota tracks published model scores on this metric so readers can compare state-of-the-art results across sources and model families.

Higher is better

Trust tiers for accuracyverifiedpapervendorcommunityunverified

Muted rows were not state of the art when published — an earlier or same-year result already scored better.

RankModelTrustScoreYearLinksFix
01o3
Non-API entry from src
unverified82.82026N/ALooks wrong?
02o4-mini
Non-API entry from src
unverified77.62026N/ALooks wrong?
03o1
Non-API entry from src
unverified75.72026N/ALooks wrong?
04o3-mini
Non-API entry from src
unverified74.92026N/ALooks wrong?
05o1-preview
Non-API entry from src
unverified73.32026N/ALooks wrong?
06gpt-45-preview
Non-API entry from src
unverified69.52026N/ALooks wrong?
07gpt-41
Non-API entry from src
unverified66.32026N/ALooks wrong?
08o1-mini
Non-API entry from src
unverified602026N/ALooks wrong?
09claude-35-sonnet
Non-API entry from src
unverified59.42026N/ALooks wrong?
10grok-2
Non-API entry from src
unverified562026N/ALooks wrong?
11llama-31-405b
Non-API entry from src
unverified50.72026N/ALooks wrong?
12claude-3-opus
Non-API entry from src
unverified50.42026N/ALooks wrong?
13gpt-4o
Non-API entry from src
unverified49.92026N/ALooks wrong?
14gpt-4-turbo
Non-API entry from src
unverified49.32026N/ALooks wrong?
15gemini-15-pro
Non-API entry from src
unverified46.22026N/ALooks wrong?
16llama-31-70b
Non-API entry from src
unverified41.72026N/ALooks wrong?
17gpt-4o-mini
Non-API entry from src
unverified40.22026N/ALooks wrong?
Lineage

GPQA in context.

See full reasoning benchmarks lineage →
This benchmark (1)
active2023-11
GPQA
§ 04 · Submit a result

Add to the leaderboard.

Submit a Result

Sign in to submit benchmark results for GPQA.

Sign in
← Back to Leaderboards