Compress tool outputs, logs, files, and RAG chunks before they reach the LLM. 20% fewer tokens for coding agents, 60-95% fewer tokens for JSON, same answers. Library, proxy, MCP server.
Python agentaianthropicclaude-codecompressioncontext-engineeringcontext-windowcursorfastapilangchainllmmcpopenaiprompt-engineeringproxypythonragtoken-optimizationtokenstypescript
ci: build the CI test wheel with a fast cargo profile
Tejas Chopraa08a0befa72e
diff --git a/.github/workflows/ci-fast.yml b/.github/workflows/ci-fast.ymlindex 28c9307d5..a5751fdf8 100644--- a/.github/workflows/ci-fast.yml+++ b/.github/workflows/ci-fast.yml@@ -62,10 +62,12 @@ jobs:- uses: Swatinem/rust-cache@v2with:workspaces: ". -> target"- - name: Build wheel once (maturin)+ - name: Build wheel once (maturin, fast CI profile)run: |python -m pip install --upgrade pip maturin- maturin build --release --out dist --interpreter "python${PY_VERSION}"+ # `--profile ci`: no LTO + parallel codegen-units → much faster compile+ # than `--release` (lto + codegen-units=1). Tests don't need the optimized ext.+ maturin build --profile ci --out dist --interpreter "python${PY_VERSION}"- uses: actions/upload-artifact@v4with:name: headroom-wheeldiff --git a/Cargo.toml b/Cargo.tomlindex c2c14f83a..4adfcbaa7 100644--- a/Cargo.toml+++ b/Cargo.toml@@ -106,3 +106,16 @@ gcp_auth = "0.12"strip = "symbols"lto = "thin"codegen-units = 1++# Fast-to-compile profile for CI test wheels. The shipped wheel uses+# `release` (lto + codegen-units=1) for runtime/size; CI only needs a working+# extension, so trade runtime perf for ~parallel, lto-free compilation. Used+# via `maturin build --profile ci`. Does NOT affect `--release` builds.+[profile.ci]+inherits = "release"+lto = false+codegen-units = 256+opt-level = 1+strip = "none"+debug = false+incremental = false