Unverified50% confidenceBenchmarkExact time
In Meta's internal memory tests, Muse Wonder outperformed systems such as Fable, Opus, and Astra.
1
Sources
50%
Confidence
Short-term (~14 days)
Relevance
10/12/2026
First Seen
Valid until: 10/26/2026
Sources
Related Entities
Related Claims
UnverifiedThe 82 tasks in OS World were not run under identical conditions—half of the tasks retained the original baseline scores, and the invested resources and number of attempts also differed—so the 7-percentage-point improvement cannot be entirely attributed to memory.64% similarUnverifiedMeta、NVIDIA等公司的研究表明RAG在知识密集型任务上的准确率比纯参数记忆提高20%-40%59% similarUnverifiedMuse Glimmer在Artificial Analysis第二版测试中幻觉率高达82%，知识工作和幻觉是其最大弱点58% similarUnverifiedAstra scored lower than small models such as GLM 5.3 Flash on the Artificial Analysis intelligence index, because that index is based on outdated and saturated benchmarks such as GPQA.57% similarUnverifiedThe user pointed out that these models achieve high scores on benchmarks yet frequently fail at the simplest tasks.57% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1010388API
curl https://kongchang.com/api/v1/knowledge/claims/1010388MCP
get_claim(id=1010388)