support vLLM cache salting in prefix aware scorer

[Frapschen]

What type of PR is this?
/kind feature

What this PR does / why we need it:
support vLLM cache salting in prefix aware scorer

Which issue(s) this PR fixes:
Fixes #1631

Does this PR introduce a user-facing change?:

support vLLM cache salting in prefix aware scorer

[netlify]

✅ Deploy Preview for gateway-api-inference-extension ready!

Name	Link
🔨 Latest commit	2f5ac56
🔍 Latest deploy log	https://app.netlify.com/projects/gateway-api-inference-extension/deploys/68d64841ee54ec000985df0d
😎 Deploy Preview	https://deploy-preview-1646--gateway-api-inference-extension.netlify.app
📱 Preview on mobile	Toggle QR Code... Use your smartphone camera to open QR code link.

---

To edit notification comments on pull requests, go to your Netlify project configuration.

[k8s-ci-robot]

Hi @Frapschen. Thanks for your PR.

I'm waiting for a kubernetes-sigs member to verify that this patch is reasonable to test. If it is, they should reply with /ok-to-test on its own line. Until that is done, I will not automatically test new commits in this PR, but the usual testing commands by org members will still work. Regular contributors should join the org to skip this step.

Once the patch is verified, the new status will be reflected by the ok-to-test label.

I understand the commands that are listed here.

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes-sigs/prow repository.

[nirrozenbaum]

/cc @liu-cong

[liu-cong]

Thanks, mostly nits, just want to make sure the parsing aligns with vllm API.

[liu-cong]

Just to confirm, did you test with both completion and chatcompletions request with vllm and make sure the parsing here works?

[Frapschen]

Yes. I have checked the request body definition in VLLM:

• ChatCompletionRequest: https://github.com/vllm-project/vllm/blob/392edee34a008af2453d936cb3cdbd97842984a7/vllm/entrypoints/openai/protocol.py#L425
• CompletionRequest: https://github.com/vllm-project/vllm/blob/392edee34a008af2453d936cb3cdbd97842984a7/vllm/entrypoints/openai/protocol.py#L1024

Both of them have cache_salt. So I sent the below curl:

for completion:

curl -i ${IP}:${PORT}/v1/completions -H 'Content-Type: application/json' -d '{
"model": "food-review-1",
"prompt": "Write as if you were a critic: San Francisco",
"max_tokens": 100,
"cache_salt": "Z3V2bmV3aGxza3ZubGFoZ3Zud3V3ZWZ2bmd0b3V2bnZmc2xpZ3RoZ2x2aQ==",
"temperature": 0
}'

parse result:

for chatcompletions:

curl -X POST -i ${IP}:${PORT}/v1/chat/completions \
  -H "Content-Type: application/json" \
  -d '{
        "model": "food-review-1",
        "max_tokens": 100,
        "temperature": 0,
        "cache_salt": "Z3V2bmV3aGxza3ZubGFoZ3Zud3V3ZWZ2bmd0b3V2bnZmc2xpZ3RoZ2x2aQ==",
        "messages": [
          {
            "role": "developer",
            "content": "You are a helpful assistant."
          },
          {
            "role": "user",
            "content": "Linux is said to be an open source kernel because "
          }
        ]
  }'

parse result:

[liu-cong]

/ok-to-test

[liu-cong]

/lgtm

[ahg-g]

/approve

[k8s-ci-robot]

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: ahg-g, Frapschen

The full list of commands accepted by this bot can be found here.

The pull request process is described here

Needs approval from an approver in each of these files:

• OWNERS [ahg-g]

Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment