Flight Log / repository
vllm-project/vllm — agent flight log
6 pull requests opened by AI coding agents in vllm-project/vllm (★ 96,542, Python): OpenAI Codex (OpenAI) 5 (50% merged), Claude Code (Anthropic) 1.
Flights by agent
| Agent | Flights | Merge rate | Median to merge | Burn rate |
|---|---|---|---|---|
| OpenAI Codex (OpenAI) | 5 | 50% | 3.6 d | 50% |
| Claude Code (Anthropic) | 1 | — | — | — |
By month
| Month | Flights | Merged | Burns |
|---|---|---|---|
| 2025-05 | 1 | 0 | 0 |
| 2025-06 | 3 | 1 | 1 |
| 2025-07 | 2 | 1 | 1 |
Where agents burned here
- OpenAI Codex (OpenAI) burn #21135
I think it would be cleaner to flip the default and then use `argparse.BooleanOptionalAction`
- OpenAI Codex (OpenAI) burn #19145
## Code Review This pull request adds `pandas` and `datasets` as dependencies to support the benchmark module, which is a useful addition. The changes are
Last 6 flights
| PR | Title | Agent | Opened | Lines | |
|---|---|---|---|---|---|
| #21146 | [Core] Freeze gc during cuda graph capture to speed up init | OpenAI Codex | Merged after rework | Jul 17, 2025 | +10,535 −15,819 |
| #21135 | [Misc] change default request logging behavior to off | OpenAI Codex | Burn | Jul 17, 2025 | +31 −2 |
| #19396 | Consolidate MoE quantization parameters into FusedMoeQuantConfig | Claude Code | Stale | Jun 10, 2025 | +487 −184 |
| #19145 | [Build] Add Benchmark dependencies to requirements/common.txt | OpenAI Codex | Burn | Jun 4, 2025 | +2 −0 |
| #19117 | [Bugfix] Fix port handling in make_zmq_path | OpenAI Codex | Merged clean | Jun 4, 2025 | +1 −1 |
| #18317 | Update arch overview for v1 | OpenAI Codex | Stale | May 18, 2025 | +70 −15 |
No rider has claimed a flight in this repository yet. Flew one of these? Open your flight page and claim it.
Read next
Data from public GitHub activity; this page is not affiliated with GitHub or the repository. Repository owner? Request exclusion — handled within 48 hours.