以下部分引用 AI 总结,现阶段 AI 仍然有幻觉。请以内容的原文为准。
数据新鲜度群聊 07-14 ✓卖方 07-14 ✓Wrap 断档15天 ✕News 07-15 ✓X 断档12天 ✕页面生成 07-15
mini-swe-agent
5 atoms · 跨 2 天 · 首见 2026-05-27 · 最近 2026-06-12
三色: 🟦 fact 5 · 🟥 take 0 · stance ▲1/▼0/◆2
来源: X 5
时态: fresh:5
标签: 好信源:3 · 好数字:2 · 好观点:1
叙事 (narrative) (3)
- 2026-06-12 · 被用作Ramp SWE-bench的评估引擎
they use mini-swe-agent as the harness powering the eval
tags: 好信源 · → daily
- 2026-06-12 · 已被移植到opencode ui
we ported mini-swe-agent to opencode ui and it's actually very nice!
tags: 好信源 · → daily
- 2026-06-12 · 在任务上表现与Claude Code和Codex一样好
showed that it performs just as well as Claude Code and Codex on the tasks
tags: 好观点 · → daily
事实 (fact) (2)
- 2026-05-27 · performance vs Claude Code/Codex/Gemini CLI on coding benchmark
mini-swe-agent is our ~150 line of code agent, and it performs as well as Claude Code, Codex and Gemini CLI on this new coding benchmark.
tags: 好数字·好信源 · → daily
value: direction=as_well_as
- 2026-05-27 · 核心agent类代码行数
mini's core agent class is just ~150 lines of code
tags: 好数字 · → daily
value: qty=~150 · direction=lines
时间轴 (近 20)
- 2026-06-12 ·
narrative · 被用作Ramp SWE-bench的评估引擎 · → daily
- 2026-06-12 ·
narrative · 在任务上表现与Claude Code和Codex一样好 · → daily
- 2026-06-12 ·
narrative · 已被移植到opencode ui · → daily
- 2026-05-27 ·
fact · 核心agent类代码行数 · → daily
- 2026-05-27 ·
fact · performance vs Claude Code/Codex/Gemini CLI on coding benchm · → daily
← 实体目录 · 系统日志