Adoption is broad. Many contributors are committing agent-authored work, which is what a real organisational rollout looks like.
Assisted or autonomous
Assisted — a developer’s identity authored the commit and the agent added itself as co-author. The agent ran git commit during a working session, with a person there. Autonomous — a bot identity is the author, no human in the loop. Across the whole index, 93.7% of agent commits are assisted.
Agent share by month
Peak month: 19.3%. Bars are scaled to that peak, not to 100%.
Which agents
| Claude Code | Anthropic | 58 | 5.4% |
| OpenAI Codex | OpenAI | 34 | 3.1% |
| GitHub Copilot | GitHub | 15 | 1.4% |
| Cursor Agent | Anysphere | 1 | 0.1% |
Verify it yourself
Sample commits that matched. Every one links to GitHub — check the trailers.
42d58c43align the precision of log2 api with torch+2.12.0 (#79541)Claude Code1dfe2f2b[Operator Mechanism] Fix p_norm_grad kernel precision compatibility for GPClaude Code31c7cfe5[Build] Compile FlashAttention on arm64 platforms (#79488)Claude Code58354a50[CI] Upgrade `apache-tvm-ffi` to `v0.1.13.post2` (#79622)OpenAI Codexc092dc58[Cpp API Compatibility] Support `at::scalar_tensor` (#79623)OpenAI Codexcae80136[Agent] Add PaddleFleet integration part and more ecosystem cases to crossOpenAI Codexcae80136[Agent] Add PaddleFleet integration part and more ecosystem cases to crossGitHub Copilot8d925a3e[Operator Mechanism] Harden MoE permute/unpermute boundary cases (#79211)GitHub Copilotd59f4f1b[CUDA13.2] Environment Adaptation support Paddle on CUDA 13.2 (#78720)GitHub Copilota5016ee4[ROCm 7.0] Add support for AMD CDNA4 and ROCm 7.0 (#77641)Cursor Agent
What was excluded
| Commits in window | 1,081 |
| Merge commits removed | −0 |
| CI / dependency bots removed | −0 |
| Considered | 1,081 |
Bots are excluded from the denominator entirely. Counting dependabot as “AI-written code” is the fastest way to produce a meaningless number.
How this was measured
Measured over February–August 2026, cloned directly. Detection is deterministic string matching against signatures agents write about themselves — no heuristics, no style analysis. Merge commits and CI bots are excluded from the denominator. The figure is a floor: squash merges strip trailers, so real usage is higher. Contributor counts are aggregate — no individual is named.
Run it yourself: npx qmmit-cli PaddlePaddle/Paddle