Claim · coding

Qwen3.8 Maxon DeepSWE

57.5 to 57.5 the range across every scoreable measurement we hold

0.0 points between the highest and the lowest

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

1 declared configuration, from 1 runner

1 / 2 rows whose cited page still shows this figure, refetched and checked by us

All 1 scoreable measurements here come from one runner under one declared configuration, so nothing on this page explains why they differ. That is the finding.

1 of 2 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.

Vendor numbers nobody has independently checked

Published by the model's own maker, on a variant of this benchmark that no independent party we hold has run. They are recorded and they are shown. They can never move a rank.

Variant	Claimed
1.1	56.6

Every measurement

2 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
57.5 / 100 as published: 57.5%	INDEPENDENT counts toward the ranking	qwen3.8-max harness mini-swe-agent effort xhigh	deepswe.datacurve.ai via Epoch AI CC-BY-4.0	checked: this figure appears in the page’s row for this model2026-08-30model:"qwen3-8-max",harness:"mini-swe-agent",reasoning_effort:"xhigh",config:"mini_swe_agent_qwen3_8_max_xhigh",source:"deep-swe",pass_rate:0.5746102449888641,p the leaderboard this came from the leaderboard for this benchmark, not a link to this row
56.6 / 100 as published: 56.6%	CLAIMED shown, never scored	qwen3.8-max harness Claude Code and mini-SWE-agent effort unknown variant 1.1	Alibaba/Qwen via Model benchmark dossier 2026-08T00:00:00.000Z facts, not expression	checked: this figure is not on the page today2026-08-30License: qwen3.8-max huggingface.co

← 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.