Unverified50% confidenceOpinionExact time
游戏实测中综合能力排名为 Sol > 5.5 > Terra > Luna
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
7/23/2026
First Seen
Valid until: 10/21/2026
Sources
GPT-5.6三模型实测对比：Sol/Terra/Luna选哪个？
bilibili麦冬AI实验室7/11/2026
Related Claims
Unverified选型建议：前端/视觉任务选 Fable 5 或 Opus 4.8，后端代理任务选 Sol，性价比代理场景选 Luna63% similarUnverified选型原则为任务越重要越往 Sol 走，任务越轻量越往 Luna 走62% similarUnverifiedLuna定位于轻量高频任务，响应快、消耗低；Terra是平衡型日常主力；SOL能力最强但成本高、响应慢，适合复杂任务62% similarUnverifiedLuna在Artificial Analysis编程智能体指数上超过了Opus 4.861% similarUnverifiedTerra 和 Luna 能以约 1/16 的成本超越 Fable 5，Terra 在编程指数上与 Fable 打平61% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/602527API
curl https://kongchang.com/api/v1/knowledge/claims/602527MCP
get_claim(id=602527)