Claim · agents

Gemini 2.5 Flash Preview (09/2025)on Terminal-Bench 2

15.4 to 17.1 the range across every scoreable measurement we hold

1.7 points between the highest and the lowest

16.4 the number the board uses: the lower median, a real row with its own provenance

4 declared configurations, from 1 runner

0 / 4 rows whose cited page still shows this figure, refetched and checked by us

All 4 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.

4 of 4 cited pages no longer show the figure we publish. The page was refetched and split at its own structural boundaries, and no block that names this model carries this number. That does not make the measurement wrong: a leaderboard is a living page and a figure recorded when it was there can be gone by the time you follow the link. It does mean the citation, as printed, no longer supports the number, and this is the only place we know of that says so about its own sources.

Every measurement

4 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.

Score	Provenance	Configuration	Who ran it	Where to check it
17.1 / 100 as published: 17.1%	INDEPENDENT counts toward the ranking	gemini-2.5-flash-preview-09-2025 harness Mini-SWE-Agent	Terminal-Bench v2 Leaderboard via Epoch AI 2025-11-03 CC-BY-4.0	checked: this figure is not on the page today2026-08-30 Terminal-Bench v2 Leaderboard
16.9 / 100 as published: 16.9%	INDEPENDENT counts toward the ranking	gemini-2.5-flash-preview-09-2025 harness Terminus 2	Terminal-Bench v2 Leaderboard via Epoch AI 2025-10-31 CC-BY-4.0	checked: this figure is not on the page today2026-08-30 Terminal-Bench v2 Leaderboard
16.4 / 100 as published: 16.4%	INDEPENDENT counts toward the ranking	gemini-2.5-flash-preview-09-2025 harness OpenHands	Terminal-Bench v2 Leaderboard via Epoch AI 2025-11-02 CC-BY-4.0	checked: this figure is not on the page today2026-08-30 Terminal-Bench v2 Leaderboard
15.4 / 100 as published: 15.4%	INDEPENDENT counts toward the ranking	gemini-2.5-flash-preview-09-2025 harness Gemini CLI	Terminal-Bench v2 Leaderboard via Epoch AI 2025-11-04 CC-BY-4.0	checked: this figure is not on the page today2026-08-30 Terminal-Bench v2 Leaderboard

← 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.