Expired90% confidenceFactTime unknown
In the Terminal Bench 2.0 benchmark, Composer 2 scored 61.7%.
1
Sources
90%
Confidence
Medium-term (~90 days)
Relevance
5/31/2026
First Seen
Valid until: 8/29/2026(expired)
Sources
Related Entities
Related Claims
VerifiedComposer 2.5 scored 63.2% on Cursor Bench V3.1, while Opus 4.7 scored 61.6% under the default configuration and GPT 5.5 scored 59.2%.77% similarUnverifiedM2.7 scored 57% on the Terminal Bench 2 test.68% similarUnverifiedIn Artificial Analysis's coding agent index, Composer 2 scored 48 and Composer 2.5 scored 63 in the Cursor Client65% similarVerifiedIn the Terminal Bench 2.0 test, K2.6 scored 66.7, slightly higher than both GPT-5.4 and Claude Opus 4.6.64% similarUnverifiedIn a landing page generation test, Composer 2.5 underperformed compared to GPT 5.5 and Opus 3.7, with a perceptible quality gap in terms of creativity and aesthetics.62% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/14073API
curl https://kongchang.com/api/v1/knowledge/claims/14073MCP
get_claim(id=14073)