Expired50% confidenceFactExact time
This comparative review adopts a five-dimension scoring system (onboarding speed, task capability, ecosystem maturity, controllability/portability, and quota cost-effectiveness), with each dimension scored out of 5 stars.
1
Sources
50%
Confidence
Short-term (~14 days)
Relevance
9/7/2026
First Seen
This claim has expired. It is kept for the record and no longer describes the current state.
Valid until: 9/21/2026(expired)
Sources
Related Entities
Related Claims
UnverifiedThe overall evaluation conclusion: Opus 5 is the quality benchmark but has the highest cost and time requirements; Grok 4.6 offers extreme speed and value for money but poor stability; Kimi K3 excels at cost control; GPT-5.6 has outstanding composition in some areas but lags overall in consistency70% similarUnverifiedBased on comprehensive empirical test results, the capability ranking of the four models is: Sol > GPT-5.5 > Terra > Luna.63% similarUnverified根据官方在 WANDR 基准测试上的数据，Grok 4.6 位于性能与效率的帕累托前沿63% similarUnverified跑分排名不等于所有场景的实际表现，Terra跑分接近5.5但游戏任务中5.5完成度略高61% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/873747API
curl https://kongchang.com/api/v1/knowledge/claims/873747MCP
get_claim(id=873747)