Unverified50% confidenceFactExact time
On strong models such as Claude Opus 4.5, DeReAct did not improve Pass@1 but made trajectories more evidence-complete and more compliant with constraints.
1
Sources
50%
Confidence
Long-term
Relevance
10/5/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedOn the Frontiers-WE benchmark, the gap between GLM-5.2 and Claude Opus 4.8 has narrowed to less than one point.61% similarUnverifiedEarlier models Claude Opus 4.6 and GLM-5.2 had a success rate of zero on these exploitation tasks.61% similarUnverifiedClaude Opus 5 approaches Fable-level performance on the FrontierCode 1.1 benchmark while costing only half as much61% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/975650API
curl https://kongchang.com/api/v1/knowledge/claims/975650MCP
get_claim(id=975650)