[router][ci] speed up python binding to 1.5 min (#12673)

.github/workflows/pr-test-rust.yml

@@ -13,7 +13,7 @@ on:
   workflow_dispatch:
 
 concurrency:
-  group: pr-test-rust-${{ github.ref }}
+  group: router-tests-${{ github.ref }}
   cancel-in-progress: true
 
 env:
@@ -70,7 +70,7 @@ jobs:
           python -c "import sglang_router; print('Python package: OK')"
           python -c "from sglang_router.sglang_router_rs import Router; print('Rust extension: OK')"
           python -m sglang_router.launch_router --help > /dev/null && echo "Entry point: OK"
-  unit-test-rust:
+  router-unit-tests:
     if: github.event_name != 'pull_request' || contains(github.event.pull_request.labels.*.name, 'run-ci')
     runs-on: ubuntu-latest
     steps:
@@ -133,7 +133,7 @@ jobs:
         if: always()
         run: sccache --show-stats
 
-  pytest-rust:
+  router-http-tests:
     if: github.event_name != 'pull_request' || contains(github.event.pull_request.labels.*.name, 'run-ci')
     runs-on: 4-gpu-a10
     timeout-minutes: 32
@@ -157,6 +157,16 @@ jobs:
           cache-all-crates: true
           cache-on-failure: true
 
+      - name: Cache router build
+        uses: actions/cache@v4
+        with:
+          path: |
+            sgl-router/target
+            sgl-router/.venv
+          key: ${{ runner.os }}-router-${{ hashFiles('sgl-router/Cargo.lock', 'sgl-router/pyproject.toml') }}
+          restore-keys: |
+            ${{ runner.os }}-router-
+
       - name: Install SGLang dependencies
         run: |
           sudo --preserve-env=PATH bash scripts/ci/ci_install_dependency.sh
@@ -165,18 +175,17 @@ jobs:
         run: |
           source "$HOME/.cargo/env"
           export RUSTC_WRAPPER=sccache
-          pip install -e "sgl-router"
           cd sgl-router
-          pip install maturin
-          maturin build --release --out dist --features vendored-openssl
-          pip install --force-reinstall dist/*.whl
+          python3 -m venv .venv
+          python3 -m pip install --upgrade pip maturin
+          maturin develop --release --profile ci --features vendored-openssl
 
 
       - name: Run Python unit tests
         run: |
           cd sgl-router
           source "$HOME/.cargo/env"
-          pip install pytest pytest-cov pytest-xdist
+          python3 -m pip install pytest pytest-cov pytest-xdist
           pytest -q py_test/unit --cov=sglang_router --cov-report=term-missing --cov-fail-under=80
 
       - name: Run Python integration tests
@@ -184,16 +193,17 @@ jobs:
           cd sgl-router
           source "$HOME/.cargo/env"
           # Integration tests use FastAPI/uvicorn for mock workers
-          pip install fastapi uvicorn orjson
-          pytest -q -m integration
+          python3 -m pip install fastapi uvicorn orjson
+          pytest -q py_test/integration_mock
 
       - name: Run Python E2E tests
         run: |
           bash scripts/killall_sglang.sh "nuk_gpus"
           cd sgl-router
+          source "$HOME/.cargo/env"
           python3 -m pip --no-cache-dir install --upgrade --ignore-installed blinker
-          python3 -m pip --no-cache-dir install --upgrade --break-system-packages genai-bench==0.0.2
-          pytest -m e2e -s  -vv -o log_cli=true --log-cli-level=INFO
+          python3 -m pip --no-cache-dir install --upgrade genai-bench==0.0.2
+          pytest py_test/e2e_http -s  -vv -o log_cli=true --log-cli-level=INFO
 
       - name: Upload benchmark results
         if: success()
@@ -202,7 +212,7 @@ jobs:
           name: genai-bench-results-all-policies
           path: sgl-router/benchmark_**/
 
-  pytest-rust-2:
+  router-grpc-response-api-tests:
     if: github.event_name != 'pull_request' || contains(github.event.pull_request.labels.*.name, 'run-ci')
     runs-on: 4-gpu-a10
     timeout-minutes: 32
@@ -226,6 +236,16 @@ jobs:
           cache-all-crates: true
           cache-on-failure: true
 
+      - name: Cache router build
+        uses: actions/cache@v4
+        with:
+          path: |
+            sgl-router/target
+            sgl-router/.venv
+          key: ${{ runner.os }}-router-${{ hashFiles('sgl-router/Cargo.lock', 'sgl-router/pyproject.toml') }}
+          restore-keys: |
+            ${{ runner.os }}-router-
+
       - name: Install SGLang dependencies
         run: |
           sudo --preserve-env=PATH bash scripts/ci/ci_install_dependency.sh
@@ -264,20 +284,22 @@ jobs:
           source "$HOME/.cargo/env"
           export RUSTC_WRAPPER=sccache
           cd sgl-router
-          pip install maturin
-          maturin build --release --out dist --features vendored-openssl
-          pip install --force-reinstall dist/*.whl
+          python3 -m venv .venv
+          python3 -m pip install --upgrade pip maturin
+          maturin develop --release --profile ci --features vendored-openssl
 
       - name: Run Python E2E response API tests
         run: |
           bash scripts/killall_sglang.sh "nuk_gpus"
           cd sgl-router
+          source "$HOME/.cargo/env"
           SHOW_ROUTER_LOGS=1 pytest py_test/e2e_response_api -s -vv -o log_cli=true --log-cli-level=INFO
 
       - name: Run Python E2E gRPC tests
         run: |
           bash scripts/killall_sglang.sh "nuk_gpus"
           cd sgl-router
+          source "$HOME/.cargo/env"
           SHOW_ROUTER_LOGS=1 ROUTER_LOCAL_MODEL_PATH="/home/ubuntu/models" pytest py_test/e2e_grpc -s -vv -o log_cli=true --log-cli-level=INFO
 
       - name: Cleanup Oracle Database
@@ -288,14 +310,14 @@ jobs:
 
 
   finish:
-    needs: [maturin-build-test, unit-test-rust, pytest-rust, pytest-rust-2]
+    needs: [maturin-build-test, router-unit-tests, router-http-tests, router-grpc-response-api-tests]
     runs-on: ubuntu-latest
     steps:
       - name: Finish
         run: echo "This is an empty step to ensure that all jobs are completed."
 
   summarize-benchmarks:
-    needs: pytest-rust
+    needs: router-http-tests
     runs-on: ubuntu-latest
     if: success()