Dataset: 2026-10-27-catalog-ai-v2
Ruleset: b2e8889c66f7
Scanned: 2026-10-27
Window: February–September 15, 2026

Catalog tracking observation. This repository is measured with the same fixed window and ruleset, but it is not part of the published benchmark or its aggregate claims.

A streamlined and customizable framework for efficient large model (LLM, VLM, AIGC) evaluation and performance benchmarking.

01Result

modelscope/evalscope ↗

modelscope/evalscope AI coding statistics: 4.8%

of 377 commits in February–September 15, 2026 were committed by an AI agent — a floor, not an estimate

4.8% agent · top contributor 33.3%11 contributors

org-wide

Adoption is broad. Many contributors are committing agent-authored work, which is what a real organisational rollout looks like.

agent commits

359

other considered commits

agent contributors

33.3%

largest contributor

02Drivers

What shapes this result

Mode, timing, and detected signatures show whether the result reflects broad adoption or one concentrated workflow.

Assisted or autonomous

18assisted

0autonomous

Assisted — a developer’s identity authored the commit and the agent added itself as co-author. Autonomous — a bot identity authored the commit without a human identity attached.

Agent share by month

Peak month: 18.2%. Bars are scaled to that peak, not to 100%.

Which agents

Agent	Vendor	Commits	Share
Gemini CLI	Google	9	2.4%
Claude Code	Anthropic	7	1.9%
GitHub Copilot	GitHub	4	1.1%

03Evidence

Inspect the matches

Public commit samples let you inspect the signatures behind the aggregate result.

Matched commit samples

ee252436 on GitHubfix(perf): honor dict body fields in line_by_line dataset (#1485)Gemini CLI
85debb7f on GitHubfix: Handle multi-line SSE format with id, event, and data fields (#1474)Gemini CLI
0f57de7c on GitHub[update] volcengine sandbox (#1337)Gemini CLI
bca26a42 on GitHubdocs: fix stale ModelAPI docstrings (#1634)Claude Code
82395a74 on GitHubfix(perf): generate HTML report for open-loop runs with no fixed rate (ratClaude Code
ff7feff6 on GitHubfix: support code execution scoring on Windows (#1488)Claude Code
85debb7f on GitHubfix: Handle multi-line SSE format with id, event, and data fields (#1474)GitHub Copilot
8cded2ae on GitHubadd Maritime-OCR-Bench support (#1388)GitHub Copilot
4b03567a on GitHubsupport line_by_line_oai custom dataset mode (#1342)GitHub Copilot

04Method

What entered the denominator

The result begins with commit history, removes merges and routine automation, and reports literal agent signatures as a lower bound.

Exclusion ledger

Commits in window	380
Merge commits removed	−3
CI / dependency bots removed	−0
Considered	377

Bots are excluded from the denominator entirely. Agent attribution requires a literal supported signature; no code-style inference is used.

Continue exploring

Claude Code commitsBrowse repositories with detected Claude Code signatures →GitHub Copilot commitsBrowse repositories with detected GitHub Copilot signatures →Gemini CLI commitsBrowse repositories with detected Gemini CLI signatures →Python AI coding statisticsCompare evidence across Python repositories →GitHub AI commit analyzerInspect another repository with the same evidence model →

Reproduce this observation

Run the same detector locally.

Deterministic signature matching; no code-style inference. Squash merges can remove trailers, so this result remains a measured floor.

npx qmmit-cli modelscope/evalscope

Scan another repository

Try

Month	Agent share	Considered commits
2026-02	0%	4
2026-03	2.9%	34
2026-04	18.2%	33
2026-05	7.8%	51
2026-06	5.6%	36
2026-07	6%	67
2026-08	0.9%	114
2026-09	0%	38