qwen-setup

Walk the user through setting up the Qwen Video Bridge plugin.

Step 1: Check Python

python3 --version

Require Python 3.9+. If not available, tell the user to install Python first.

Step 2: Install dashscope

pip install dashscope --break-system-packages 2>&1 | tail -3

If pip is not available, try pip3. If neither works, suggest python3 -m pip install dashscope.

Step 3: Check API key

echo "${DASHSCOPE_API_KEY:-NOT_SET}" | head -c 10

If NOT_SET, guide the user:

1. Go to https://modelstudio.console.alibabacloud.com/ (international) or https://dashscope.console.aliyun.com/ (China)
2. Sign up / log in
3. Go to API Keys section
4. Create a new key
5. Set it as an environment variable:
   • macOS/Linux: Add export DASHSCOPE_API_KEY=sk-your-key-here to ~/.bashrc or ~/.zshrc
   • Windows: Add it to System Environment Variables, or in PowerShell: $env:DASHSCOPE_API_KEY = "sk-your-key-here"

Step 4: Verify

Run a quick test with a simple prompt (no video needed — test the connection):

python3 -c "
import dashscope
dashscope.base_http_api_url = 'https://dashscope-intl.aliyuncs.com/api/v1'
import os
from dashscope import MultiModalConversation
r = MultiModalConversation.call(
    api_key=os.environ['DASHSCOPE_API_KEY'],
    model='qwen-omni-plus-latest',
    messages=[{'role':'user','content':[{'text':'Say OK if you can hear me.'}]}]
)
if r.status_code == 200:
    print('SUCCESS: Qwen API is working!')
else:
    print(f'ERROR: {r.status_code} — {r.message}')
"

If successful, tell the user they're all set and can use /analyze-video or just describe a video in conversation.

If it fails, help troubleshoot:

• 401 → invalid API key
• 429 → rate limit, wait and retry
• Connection error → check network, might need to switch to CN endpoint