[9/n] decouple quantization impl from vllm dependency - adjust ci (#12753)

.github/workflows/quantization-test.yml

@@ -1,4 +1,4 @@
-name: VLLM Dependency Test
+name: Quantization Test
 
 on:
   push:
@@ -8,7 +8,7 @@ on:
       - "!python/sglang/multimodal_gen/**"
       - "scripts/ci/**"
       - "test/**"
-      - ".github/workflows/vllm-dependency-test.yml"
+      - ".github/workflows/quantization-test.yml"
   pull_request:
     branches: [ main ]
     paths:
@@ -16,16 +16,16 @@ on:
       - "!python/sglang/multimodal_gen/**"
       - "scripts/ci/**"
       - "test/**"
-      - ".github/workflows/vllm-dependency-test.yml"
+      - ".github/workflows/quantization-test.yml"
     types: [synchronize, labeled]
   workflow_dispatch:
 
 concurrency:
-  group: vllm-dependency-test-${{ github.ref }}
+  group: quantization-test-${{ github.ref }}
   cancel-in-progress: true
 
 jobs:
-  vllm-dependency-test:
+  quantization-test:
     if: github.event_name != 'pull_request' || contains(github.event.pull_request.labels.*.name, 'run-ci')
     runs-on: 1-gpu-runner
     steps:
@@ -37,8 +37,8 @@ jobs:
           bash scripts/ci/ci_install_dependency.sh
           pip install "bitsandbytes>=0.44.0"
 
-      - name: Run vLLM dependency tests
+      - name: Run Quantization tests
         timeout-minutes: 30
         run: |
           cd test/srt
-          python3 run_suite.py --suite vllm_dependency_test
+          python3 run_suite.py --suite quantization_test

scripts/ci_monitor/ci_analyzer.py

@@ -65,7 +65,7 @@ class SGLangCIAnalyzer:
 
     def analyze_ci_failures(self, runs: List[Dict]) -> Dict:
         print(
-            "Analyzing CI failure data (pr-test.yml, vllm-dependency-test.yml, nightly-test.yml jobs only)..."
+            "Analyzing CI failure data (pr-test.yml, quantization-test.yml, nightly-test.yml jobs only)..."
         )
 
         job_categories = {
@@ -111,7 +111,7 @@ class SGLangCIAnalyzer:
             ],
             "integration": [
                 "run-all-notebooks",
-                "vllm-dependency-test",
+                "quantization-test",
                 "test-disaggregation",
             ],
             "b200": [
@@ -189,7 +189,7 @@ class SGLangCIAnalyzer:
                     "unit-test-backend-8-gpu-deepseek-v32",
                     "unit-test-backend-4-gpu-b200",
                     "unit-test-backend-4-gpu-gb200",
-                    "vllm-dependency-test",
+                    "quantization-test",
                     "nightly-test-eval-text-models",
                     "nightly-test-perf-text-models",
                     "nightly-test-eval-vlms",

test/srt/run_suite.py

@@ -199,11 +199,11 @@ suites = {
         TestFile("test_deepseek_v32_mtp.py", 275),
         TestFile("test_deepseek_v32_nsabackend.py", 600),
     ],
-    "vllm_dependency_test": [
+    "quantization_test": [
         TestFile("quant/test_awq.py", 163),
         TestFile("test_bnb.py", 5),
         TestFile("test_gptqmodel_dynamic.py", 102),
-        TestFile("test_vllm_dependency.py", 185),
+        TestFile("test_quantization.py", 185),
         TestFile("test_gguf.py", 96),
     ],
     # If the test cases take too long, considering adding them to nightly tests instead of per-commit tests

test/srt/test_quantization.py

@@ -77,7 +77,7 @@ def check_model_scores(results):
 
     if is_in_ci():
         write_github_step_summary(
-            f"### TestNightlyGsm8KEval for vLLM awq, gptq, gguf\n{summary}"
+            f"### TestNightlyGsm8KEval for awq, gptq, gguf\n{summary}"
         )
 
     if failed_models: