Unverified50% confidenceBenchmarkExact time
In both max thinking mode and the Full version, MuseWonder achieved benchmark scores higher than the previous generation MuseSpark 1.3 max with lower token consumption.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/13/2026
First Seen
Valid until: 1/11/2027
Sources
Related Entities
Related Claims
UnverifiedMuse-Glimmer-30B在智能体工作流得90.87分、指令遵循90.86分、工具/CLI 88.84分，三项均优于Corpus 3.6 27B69% similarUnverifiedMuse Glimmer在DeepSearch QA得分74.6，SWE-Bench Pro得分51.268% similarUnverifiedMuse-Glimmer-30B surpasses the Qwen-3 series 27B model on some benchmark metrics, but fails to surpass it on others.68% similarUnverifiedMuse Spark 1.3 achieved an accuracy rate of 98.1% in ultra-long context tests of one million tokens.68% similarUnverifiedOn DeepSWE 1.1, Muse Wonder beats Claude Opus 5.5 78.8 to 74.2, leads 10.0 to 4.2 on the Harvey legal agent benchmark LAB-AA, and leads 53.1 to 46.3 on SWE-Marathon.67% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1014691API
curl https://kongchang.com/api/v1/knowledge/claims/1014691MCP
get_claim(id=1014691)