fix: set env in runner (#891)

.github/workflows/e2e-test.yml

@@ -21,26 +21,24 @@ jobs:
   e2e-test:
     runs-on: self-hosted
 
-    env:
-      HF_TOKEN : ${{ secrets.HF_TOKEN }}
-
     steps:
     - name: Checkout code
       uses: actions/checkout@v3
 
     - name: Install dependencies
       run: |
-        cd /data/zhyncs/venv && source ./bin/activate && cd -
+        source $HOME/venv/bin/activate
+        echo "$HOME/venv/bin" >> $GITHUB_PATH
 
         pip cache purge
         pip install --upgrade pip
         pip install -e "python[all]"
         pip install flashinfer -i https://flashinfer.ai/whl/cu121/torch2.3/ --force-reinstall
         pip install --upgrade transformers
+        pip install python-multipart
 
     - name: Benchmark Serving Throughput
       run: |
-        cd /data/zhyncs/venv && source ./bin/activate && cd -
         python3 -m sglang.launch_server --model meta-llama/Meta-Llama-3.1-8B-Instruct --port 8413 --disable-radix-cache &
 
         echo "Waiting for server to start..."
@@ -56,7 +54,7 @@ jobs:
           sleep 1
         done
 
-        cd /data/zhyncs && python3 -m sglang.bench_serving --backend sglang --port 8413 --dataset-name random --num-prompts 3000 --random-input 256 --random-output 512
+        cd $HOME && python3 -m sglang.bench_serving --backend sglang --port 8413 --dataset-name random --num-prompts 3000 --random-input 256 --random-output 512
 
         echo "Stopping server..."
         kill -9 $(ps aux | grep sglang | grep Meta-Llama-3.1-8B-Instruct | grep -v grep | awk '{print $2}')

.github/workflows/unit-test.yml

@@ -21,16 +21,14 @@ jobs:
   unit-test:
     runs-on: self-hosted
 
-    env:
-      HF_TOKEN : ${{ secrets.HF_TOKEN }}
-
     steps:
     - name: Checkout code
       uses: actions/checkout@v3
 
     - name: Install dependencies
       run: |
-        cd /data/zhyncs/venv && source ./bin/activate && cd -
+        source $HOME/venv/bin/activate
+        echo "$HOME/venv/bin" >> $GITHUB_PATH
 
         pip cache purge
         pip install --upgrade pip
@@ -40,29 +38,20 @@ jobs:
 
     - name: Test Frontend Language with OpenAI Backend
       run: |
-        cd /data/zhyncs/venv && source ./bin/activate && cd -
-        export OPENAI_API_KEY=${{ secrets.OPENAI_API_KEY }}
-
         cd test/lang
         python3 test_openai_backend.py
 
     - name: Test Frontend Language with SRT Backend
       run: |
-        cd /data/zhyncs/venv && source ./bin/activate && cd -
-
         cd test/lang
         python3 test_srt_backend.py
 
     - name: Test OpenAI API Server
       run: |
-        cd /data/zhyncs/venv && source ./bin/activate && cd -
-
         cd test/srt
         python3 test_openai_server.py
 
     - name: Test Accuracy
       run: |
-        cd /data/zhyncs/venv && source ./bin/activate && cd -
-
         cd test/srt
         python3 test_eval_accuracy.py

test/lang/test_bind_cache.py

@@ -1,6 +1,7 @@
 import unittest
 
 import sglang as sgl
+from sglang.test.test_utils import MODEL_NAME_FOR_TEST
 
 
 class TestBind(unittest.TestCase):
@@ -8,7 +9,7 @@ class TestBind(unittest.TestCase):
 
     @classmethod
     def setUpClass(cls):
-        cls.backend = sgl.Runtime(model_path="meta-llama/Meta-Llama-3-8B-Instruct")
+        cls.backend = sgl.Runtime(model_path=MODEL_NAME_FOR_TEST)
         sgl.set_default_backend(cls.backend)
 
     @classmethod