Trending storyPast story
Unreal Agent: async harness cuts agent costs by 40% on GPT-6 Astra
1 report1 sourceupdated 7 days ago
What happened
Summary
Unreal Labs 开源了一个 agent 调度器，核心思路是把工具调用改成全异步：模型发出指令后不用干等，工具在后台跑，结果回来再补进对话记录。在 Terminal-Bench、SWE-Atlas、DeepSWE 和 ALE-CLI 四个基准上，用 GPT-6 Astra xhigh 跑，成本比 Codex 最多低 40%，比 Pi 最多低 20...
Coverage
Follow the reports to see the story from different sides.
Sep 23
- Hacker News front pageUnreal Agent: async harness cuts agent costs by 40% on GPT-6 Astra
Unreal Labs open-sourced an agent harness that makes tool calls fully asynchronous: the model issues a call and moves on while the tool runs in the background, with results appended later. On Terminal-Bench, SWE-Atlas, DeepSWE, and ALE-CLI with GPT-6 Astra xhigh, it costs up to 40% less than Codex and up to 20% less than Pi, with pass rates roughly equal. The post doesn't report latency numbers or results with non-GPT-6 models.