CursorBench 3.1

Claude Fable 5 72.9% leads since Jun 2026 Fading

Agents do ambiguous multi-file coding tasks from real Cursor sessions in Cursor's harness. Version 3.1 added codebase understanding, bug finding, planning and code review problems. Agentic graders score correctness.

Cursor's changelog dates it 2026-05-19 and says it also improved grading for some edit tasks. Cursor says 3.1 scores should be compared only within the same version.

Made by Cursor Released May 2026 Fixing real repositories Tasks kept private Last in a launch post May 2026 , dropped for a newer version Last model added Jun 2026

Level: Expert Specialists with years of training, often over hours of work
Scored: Graded by a model A language model scores each response, usually against a rubric.
Answers: Code A program or patch run against tests.

Leaderboard cursor.com

Models scored in lab cards

Launch posts list it, from 1 lab

System cards report it, from 1 lab

Claude Fable 5 leads with 72.9% among models released by Jun 2026

Best score per model as labs report them in their own cards. Labs pick their own settings, so these compare less cleanly than independent runs.

Bars Table

The newest model with a score here came out in Jun 2026, and nobody has reported a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:

1 Claude Fable 5 72.9%
2 Composer 2.5 63.2%
3 Composer 2 61.3%
4 Claude Sonnet 5 61.2%

#	Model	Score	Date	Source
1	Claude Fable 5	72.9%	Jun 2026	Model card
2	Composer 2.5	63.2%	May 2026	Model card
3	Composer 2	61.3%	Mar 2026	Model card
4	Claude Sonnet 5	61.2%	Jun 2026	Model card

3 models have held number one since its release in May 2026

Lab claims count from the day the card came out, independent runs from the model’s release.

Cursor Anthropic

Lab claims Replaced by 3.2, Jul 2026

Claude Fable 5 72.9%

Scores over time

All the scores we have, one dot each, so a model tested at several settings appears more than once.

Reported by the lab in its model card Best so far

Labs headline it less often now, last in May 2026

Launch posts from every lab that has listed it, one tick per post. Tall colored ticks include it, and short gray ones leave it out.

Cursor

Date	Launch post	Score	As the post writes it
18 May 2026	Composer 2.5	63.2	CursorBench v3.1 (harder tasks)
19 Mar 2026	Composer 2	61.3	CursorBench

2 system cards report it, from 1 lab

Besides launch posts, we read the system cards, model cards and technical reports labs publish with a release. They report far more, including safety and frontier-risk tests.

Date	Document	Score	As the document writes it	Section
30 Jun 2026	Claude Sonnet 5 system card	61.2	CursorBench at max effort, measured by Cursor, version not stated (xhigh 59.0%, high 57.0%) (8.5, Figure 8.5.A)	Capabilities
9 Jun 2026	Claude Fable 5 system card	72.9	CursorBench at max effort, measured and reported by Cursor in its production harness, version not stated (8.7, Figure 8.7.A)	Capabilities

Versions

We switch the top row to each new version when it comes out. Open it for one row per version. An old version stays in play while people still test it, so whoever first takes it past 85% gets credit for closing it out.

Cursor xAI Anthropic

CursorBench 4.0 5 leaders