DEDUPEDWho it is forEngineer evaluating gitai who does not want to trust published numbers without running them
The problemThe README claims 200 landed changes versus about 100 for classic landing, but you need to see it on your hardware.
The outcomeYou get your own landed-count, attempts-per-change and latency figures for 50 agents landing 250 changes with planned conflicts.
What you need first
- Linux or macOS
- Go 1.26, a C toolchain, Meson and Ninja (the build compiles a bundled Git)
- No paid keys needed; conflict resolution can use any LLM provider, a deterministic tool, or none
- SQLite or PostgreSQL plus object storage (file, S3, GCS, Azure) only if you move the repo state off local Git
Steps
- Build the repo with: make build
- Run the benchmark script at examples/graft/demo/acceptance-phase4.sh
- Compare the classic and gitai columns for changes landed and attempts per landed change
- Read doc/graft/BENCH.md for the other agent counts
Prompt for your agent
In feder-cr/git-ai, run examples/graft/demo/acceptance-phase4.sh after make build and report changes landed, attempts per landed change and p99 latency for classic versus gitai. Verify that the roughly 50 unlanded changes in the gitai run are the planned conflicts.
Caveat. The README numbers come from one 16-core machine and a run without a conflict resolver; a smaller machine will show different lands per minute.
About git-ai
Version control built for coding agents instead of humans, so many agents can work on one codebase without stepping on each other. View feder-cr/git-ai on GitHub.
More with git-ai
- Give every coding agent its own token and branch limits Use per-agent tokens, an RPC allowlist and branch policies so each agent can only do what you allow.
- Watch many agents land changes on one branch in minutes Build gitai and run the bundled quickstart to see agents land on a shared branch with no merge queue.
- Stand up a shared server for a fleet of coding agents Run your own gitai server so each agent gets its own token and lands onto shared branches.
- Resolve conflicting agent changes automatically with a chosen resolver Plug in a deterministic tool or any LLM provider so conflicting lands resolve instead of bouncing back to the agent.
- Give an agent only the folders it needs in a huge repo Use graft sparse and graft clone to materialize just the paths a short task touches instead of a full clone.
- Test the branch tip after landing and revert the culprit Run graft-verify so a failing tip is bisected and the change that broke it is reverted without a human.
- Follow the live changeset log with a monitoring agent Have an agent watch every accepted update and report what matters to each component, including in Slack.
- Move repo state into PostgreSQL and object storage for scale-out Switch a gitai repo to a database plus object storage so any server node can be added or rebuilt.
Disclaimer
Deduped lists open-source repositories that belong to other people. We do not own, maintain, audit, or vouch for any of them, and a repo appearing here is not an endorsement or a safety check. The use cases, steps, and prompts are written by AI and may be incomplete, outdated, or just wrong. Where a use case is marked "Seen in the wild", we link to public posts where someone described something similar, but we have not verified those posts and the use case itself is still our AI's own write-up. Use all of this at your own risk. Only run code, commands, or agents on systems and accounts you are authorized to use, and read the code and every command before you run it, especially anything that touches credentials, files, or the network. Nothing here is professional, security, legal, or financial advice. All trademarks and project names belong to their respective owners, and their use here does not imply affiliation. Content is provided as is, without warranties of any kind. Found a mistake, a safety problem, or something you want removed? Email [email protected] and we will look at it promptly. Full disclaimer page