Update vLLM version to v0.9.0

[CICD-at-OPEA]

Update vLLM version to v0.9.0

[github-actions[bot]]

Dependency Review

✅ No vulnerabilities or license issues found.

Scanned Files

None

[chensuyue]

https://github.com/opea-project/GenAIExamples/issues/1933

[eero-t]

Regarding CI fail, I don't think vLLM used for embeddings yet, in which case this would be unrelated to vLLM update:

+ echo '[ mega-codegen ] HTTP status is not 200. Received status was 500'
+ docker logs codegen-xeon-backend-server
...
   File "/usr/local/lib/python3.11/site-packages/aiohttp/client_reqrep.py", line 1281, in json
    raise ContentTypeError(
aiohttp.client_exceptions.ContentTypeError: 500, message='Attempt to decode JSON with unexpected mimetype: text/plain; charset=utf-8', url='http://10.244.47.206:6000/v1/embeddings'
+ exit 1

[chensuyue]

Regarding CI fail, I don't think vLLM used for embeddings yet, in which case this would be unrelated to vLLM update:

+ echo '[ mega-codegen ] HTTP status is not 200. Received status was 500'
+ docker logs codegen-xeon-backend-server
...
   File "/usr/local/lib/python3.11/site-packages/aiohttp/client_reqrep.py", line 1281, in json
    raise ContentTypeError(
aiohttp.client_exceptions.ContentTypeError: 500, message='Attempt to decode JSON with unexpected mimetype: text/plain; charset=utf-8', url='http://10.244.47.206:6000/v1/embeddings'
+ exit 1

Yes, I paste the issue before, #1933

[chensuyue]

Looks like in v0.9.0 release torch version doesn't match intel_extension_for_pytorch version, not sure if there is the main issue. https://github.com/vllm-project/vllm/blob/dc1440cf9f8f6233a3c464e1a01daa12207f8680/requirements/cpu.txt#L6 https://github.com/vllm-project/vllm/blob/dc1440cf9f8f6233a3c464e1a01daa12207f8680/docker/Dockerfile.cpu#L55

ERROR! Intel® Extension for PyTorch* needs to work with PyTorch 2.6.*, but PyTorch 2.7.0+cpu is found. Please switch to the matching version and run again.

[chensuyue]

We may need to wait for a new release or use v0.8.5.

[chensuyue]

Looks like in v0.9.0 release torch version doesn't match intel_extension_for_pytorch version, not sure if there is the main issue. https://github.com/vllm-project/vllm/blob/dc1440cf9f8f6233a3c464e1a01daa12207f8680/requirements/cpu.txt#L6 https://github.com/vllm-project/vllm/blob/dc1440cf9f8f6233a3c464e1a01daa12207f8680/docker/Dockerfile.cpu#L55

ERROR! Intel® Extension for PyTorch* needs to work with PyTorch 2.6.*, but PyTorch 2.7.0+cpu is found. Please switch to the matching version and run again.

The latest release v0.9.0.1 solve this issue, https://github.com/vllm-project/vllm/commit/7ab056c273dbf89ac1e157e539a8b2c1967fb263

[chensuyue]

AudioQnA issue trace in https://github.com/opea-project/GenAIExamples/issues/2048