Benchmarks

Measured by scripts/bench.py in the bench workflow on a standard GitHub-hosted ubuntu-latest runner. Anyone can re-run it.

Run 36094170937 (repomap 1.0.0, 4 vCPU ubuntu-latest):

Repository	Code files	Symbols	Resolved calls	Cold index	Warm index	Peak RSS
kubernetes (Go)	11,710	101,176	200,157	11.2 s	1.88 s	456 MB
vscode (TypeScript)	6,125	95,785	182,506	8.0 s	1.35 s	320 MB
django (Python)	2,271	41,019	59,643	2.2 s	0.42 s	125 MB
rust-analyzer (Rust)	1,512	29,024	59,438	1.8 s	0.31 s	101 MB

Query latency over MCP stdio, p50 / p95:

Repository	search	context	impact	trace
kubernetes	57.3 / 68.2 ms	0.8 / 4.2 ms	5.3 / 9.8 ms	0.4 / 0.5 ms
vscode	6.6 / 47.6 ms	1.6 / 2.2 ms	6.4 / 9.0 ms	0.3 / 0.3 ms
django	20.9 / 26.0 ms	0.3 / 1.9 ms	0.8 / 3.2 ms	0.2 / 0.3 ms
rust-analyzer	3.6 / 17.3 ms	0.4 / 1.1 ms	1.7 / 2.9 ms	0.3 / 0.3 ms

Method

• Corpora: shallow clones at fixed tags: microsoft/vscode 1.104.0 (TypeScript), rust-lang/rust-analyzer 2026-09-21 (Rust), django/django 5.2.6 (Python) and kubernetes/kubernetes v1.34.1 (Go).
• Cold index: repomap index <repo> --no-cache, the median of 3 runs. This covers the walk, tree-sitter parse, import and call resolution, PageRank, Louvain and BM25.
• Warm index: the same command with the per-file cache populated. This is what a restart of the MCP server costs.
• Peak RSS: the highest resident memory of the index runs.
• Query latency: one long-lived repomap mcp process, with the round trip measured over stdio JSON-RPC. search runs 5 phrases plus the 10 most used symbol names. context, impact and trace (callers) each run on those 10 symbols. The table shows p50; p95 is in the JSON artifact.

Versus GitNexus

We have not run GitNexus in this benchmark. Its PolyForm Noncommercial licence does not allow use for a company's commercial purposes, and a vendor benchmark could count as one. Its README documents its own performance. You are welcome to run both tools on the same corpora on your own machine.