How to Deploy Qwen3.5-35B-A3B with vLLM

# Create a working directory (choose your preferred location)

# Create virtual environment with Python 3.12
python3.12 -m venv venv

# Activate virtual environment
source venv/bin/activate

# Verify Python version (should be 3.12.x)
python --version

# Upgrade pip
pip install --upgrade pip

# Install vLLM nightly (REQUIRED — Qwen3.5 is not yet in stable vLLM)
# Option A: Using uv (faster)
pip install uv
uv pip install vllm --torch-backend=auto --extra-index-url https://wheels.vllm.ai/nightly

# Option B: Using pip directly
pip install vllm --extra-index-url https://wheels.vllm.ai/nightly

# Install video support
pip install 'vllm[video]'

# Make sure you're in your working directory and activate virtual environment
source venv/bin/activate

# Launch vLLM serve
vllm serve Qwen/Qwen3.5-35B-A3B \
  --dtype bfloat16 \
  --max-model-len 16384 \
  --max-num-batched-tokens 16384 \
  --gpu-memory-utilization 0.90 \
  --port 8000

curl http://localhost:8000/v1/models

curl http://localhost:8000/v1/completions \
  -H "Content-Type: application/json" \
  -d '{
    "model": "Qwen/Qwen3.5-35B-A3B",
    "prompt": "Hello, how are you?",
    "max_tokens": 50
  }'

VIDEO_BASE64=$(base64 -w 0 /path/to/your/video.mp4)

time curl http://localhost:8000/v1/chat/completions \
  -H "Content-Type: application/json" \
  -d "{
    \"model\": \"Qwen/Qwen3.5-35B-A3B\",
    \"messages\": [{
      \"role\": \"user\",
      \"content\": [
        {
          \"type\": \"video_url\",
          \"video_url\": {
            \"url\": \"data:video/mp4;base64,\$VIDEO_BASE64\"
          }
        },
        {
          \"type\": \"text\",
          \"text\": \"Describe the user actions in this video in a sequential list.\"
        }
      ]
    }],
    \"max_tokens\": 1024
  }"