Unverified50% confidenceBenchmarkExact time
The Luna model's fail_fast/endstate score is 64/55, with the gap corresponding to trajectories that appear clean turn by turn but end in a corrupted final state.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
9/22/2026
First Seen
Valid until: 12/21/2026
Sources
Related Entities
Related Claims
UnverifiedLuna scored 53.74 points on the backend and 7.32 points on the frontend, with frontend interaction being largely non-functional.67% similarUnverifiedOn the Deep Squeeze test, the old model scored 49%, while Gemini 3.7 Flash jumped to 65.3%.64% similarUnverifiedLuna配置在HIGH和XH模式下10次测试全部答错，正确率为零63% similarUnverifiedIn actual testing, the Top Speed configuration completed the run in 14 minutes consuming 2 million tokens; the Kimi K3 model in the Open Weights configuration failed entirely and could not output correct JSON; and the planning phase of Opus 5 in the Frontier configuration ran for 13 minutes.61% similarUnverifiedIn the Terminal Bench test, Qwen 3.6 27B scored approximately 60.7, while Muse Glimmer scored only 51.7, falling significantly behind.58% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/941205API
curl https://kongchang.com/api/v1/knowledge/claims/941205MCP
get_claim(id=941205)