Unverified50% confidenceBenchmarkExact time
On ProgramBench, Muse Wonder scored 50.0 and Claude Opus 5.5 scored 70.5, with Opus ahead by 20.5 points, the largest gap among the 14 benchmarks.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/12/2026
First Seen
Valid until: 1/10/2027
Sources
Related Entities
Related Claims
UnverifiedIn the Intelligence Index given by the independent evaluation organization Artificial Analysis, Muse Wonder scores only 52, while Claude Opus 5.5 reaches 58.73% similarUnverifiedThe model scored 51 points on the Agentic Index, surpassing Claude Opus's highest reasoning tier, but fell behind the Claude Opus level on Terminal Bench and GPQA Diamond.67% similarUnverifiedClaude Opus 5 achieved a pass@20 of 79.09% on ThinkingBox, with an all-20 of 47.53% (241 tasks passing consistently).66% similarUnverifiedMeta's chart shows Muse Wonder beating Claude Opus 5.5 on long-horizon coding tasks: DeepSWE score 78.8 versus 74.2, SWE-Marathon score 53.1 versus 46.3.66% similarUnverifiedOn the Harvey LAB-AA legal benchmark, Muse Wonder scored 10.0 and Claude Opus 5.5 scored 4.2, doubling Wonder's score, but the highest score in the whole row is only Wonder Full's 13.1.65% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1011248API
curl https://kongchang.com/api/v1/knowledge/claims/1011248MCP
get_claim(id=1011248)