Claim · reasoning

GPT-5.4 mini (2026-03-17)on GPQA Diamond (Epoch's own run)

64.1 to 86.9 the range across every scoreable measurement we hold
22.7 points between the highest and the lowest
83.6 the number the board uses: the lower median, a real row with its own provenance
3 declared configurations, from 1 runner

All 3 scoreable measurements here come from ONE runner, so this is not a reconciliation and is not presented as one. What it shows is how far a single runner's number moves when the model is run differently.

Every measurement

3 rows. Best first. the board takes the LOWER MEDIAN of the scoreable measurements, which is a real row with its own interval and provenance rather than an interpolated average. a vendor self-report is shown beside the independent numbers and is never one of them; it cannot move a rank.

ScoreProvenanceConfigurationWho ran itWhere to check it
86.9 / 100
as published: 86.9%
INDEPENDENT
counts toward the ranking
gpt-5.4-mini-2026-03-17
effort xhigh
Epoch AI
2026-08-07T01:00:11.000Z
CC-BY-4.0
NO_URL2026-08-30
benchmark leaderboard
the leaderboard for this benchmark, not a link to this row
83.6 / 100
as published: 83.6%
INDEPENDENT
counts toward the ranking
gpt-5.4-mini-2026-03-17
effort high
Epoch AI
2026-04-15T14:03:19.000Z
CC-BY-4.0
NO_URL2026-08-30
benchmark leaderboard
the leaderboard for this benchmark, not a link to this row
64.1 / 100
as published: 64.1%
INDEPENDENT
counts toward the ranking
gpt-5.4-mini-2026-03-17
effort none
Epoch AI
2026-08-07T00:46:46.000Z
CC-BY-4.0
NO_URL2026-08-30
benchmark leaderboard
the leaderboard for this benchmark, not a link to this row
← The Forge Rankings

Snapshot 2026-08-30-62e4e85f043b, manifest dc46131315046f1e, as of 2026-08-30. where only one party has measured a model on a benchmark, this page says so rather than presenting a single number as a reconciliation.