Primary sourceArtificial Analysis
View sourceEvent ·Multi-source
GitHub 发布 Copilot Fusion,动态调度 Anthropic 与 OpenAI 模型
GitHub 在 GitHub Copilot 中推出 Copilot Fusion,一个运行时编排系统,会在 Claude Opus 5.5、Claude Sonnet 5.5、GPT-6 Astra、GPT-6 Sol 等模型间自动选择,以优化性能、成本和延迟。在 Artificial Analysis Coding Agent Index v2.0 中,其 max 设置每任务成本为 5.59 美元,仅为 Claude Code 用 Opus 5.5(11.42 美元)的一半左右,成绩低 2 分。它在 Terminal-Bench 4.0 拿到 71 分排名第一,FrontierCode 得 51 分排第三,每任务消耗 7.2M token,约为 Claude Code 中 Opus 5.5 的一半。Token 费用按实际调用的底层模型计费。
Current read
GitHub 新出的 Copilot Fusion 会自动在 Claude 和 GPT 模型里挑着用,跑分只比 Opus 5.5 低 2 分,成本却便宜一半。
11 sources72° AI heatLast updated 2026/10/28 16:14:44
Evidence
11 sourcesRelated source 1TestingCatalog
View sourceRelated source 2The Rundown AI
View sourceRelated source 3IT之家
View sourceRelated source 4@koltregaskes
View sourceRelated source 5Yuchen Jin
View sourceRelated source 6Claude Code: GitHub Releases
View sourceRelated source 7Latent Space (swyx)
View sourceRelated source 8kimmonismus
View sourceRelated source 9Simon Willison’s Weblog
View sourceRelated source 10Bindu Reddy
View sourceConflict check
Dedup data only shows that these sources discuss the same event, not that they agree. There is no structured supporting or refuting evidence yet, so no conflict is inferred automatically.