Claim · agents

Jedi 7B W/ GPT-4oon OSWorld

26.8 to 29.3 the range across every scoreable measurement we hold

2.5 points between the highest and the lowest

27.0 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.

What the vendor said, and what someone else measured

Matched on which variant of the benchmark was sat, and on nothing else: the vendor ran it on their harness and the independent party on theirs, and that difference is the comparison rather than an obstacle to it. A vendor number with no independent run of the same variant carries no delta, because a number that cannot be compared honestly must not be compared.

Variant	Vendor claims	Independently measured	Difference
as published, no variant stated	27.0	27.0 from 3 runs	+0.0

Every measurement

6 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
29.3 / 100 as published: 29.3%	INDEPENDENT counts toward the ranking	Jedi-7B w/ gpt-4o harness Agentic framework effort 100 steps	OSWorld (XLANG Lab) via OSWorld leaderboard none stated	NO_URL2026-08-30 benchmark leaderboard the leaderboard for this benchmark, not a link to this row
27.0 / 100 as published: 27.0%	CLAIMED shown, never scored	Jedi-7B w/ GPT-4o (100 steps)	OSWorld (XLANG Lab) via OSWorld leaderboard none stated	NO_URL2026-08-30 benchmark leaderboard the leaderboard for this benchmark, not a link to this row
27.0 / 100 as published: 27.0%	INDEPENDENT counts toward the ranking	Jedi-7B w/ gpt-4o harness Agentic framework effort 50 steps	OSWorld (XLANG Lab) via OSWorld leaderboard none stated	NO_URL2026-08-30 benchmark leaderboard the leaderboard for this benchmark, not a link to this row
26.8 / 100 as published: 26.8%	INDEPENDENT counts toward the ranking	Jedi-7B w/ gpt-4o harness Agentic framework effort 15 steps	OSWorld (XLANG Lab) via OSWorld leaderboard none stated	NO_URL2026-08-30 benchmark leaderboard the leaderboard for this benchmark, not a link to this row
25.0 / 100 as published: 25.0%	CLAIMED shown, never scored	Jedi-7B w/ GPT-4o (50 steps)	OSWorld (XLANG Lab) via OSWorld leaderboard none stated	NO_URL2026-08-30 benchmark leaderboard the leaderboard for this benchmark, not a link to this row
22.7 / 100 as published: 22.7%	CLAIMED shown, never scored	Jedi-7B w/ GPT-4o (15 steps)	OSWorld (XLANG Lab) via OSWorld leaderboard none stated	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.