fix(reranker): llama-cpp sort score desc, crop top_n

[mkhludnev]

Description

This PR fixes #7189

Notes for Reviewers

After applying the fix I see the right reranker output:

curl -X POST http://localhost:8080/v1/rerank \
  -H "Content-Type: application/json" \
  -d '{
    "query": "What are the benefits of exercise?",
    "documents": [
      "Regular exercise can improve cardiovascular health.",
      "Eating a balanced diet is important for overall well-being.",
      "Exercise helps in weight management and builds muscle strength.",
      "Reading books can expand your knowledge and vocabulary."
    ],
    "model": "jina-reranker-v1-tiny-en",
    "top_n": 2
  }'


{
  "model": "jina-reranker-v1-tiny-en",
  "usage": {
    "total_tokens": 41,
    "prompt_tokens": 41
  },
  "results": [
    {
      "index": 2,
      "document": {
        "text": "Exercise helps in weight management and builds muscle strength."
      },
      "relevance_score": 0.07032939791679382
    },
    {
      "index": 0,
      "document": {
        "text": "Regular exercise can improve cardiovascular health."
      },
      "relevance_score": 0.059670690447092056
    }
  ]
}

is there are tests, which I can assert behavior for this backend?

Signed commits

• [v] Yes, I signed my commits.

[netlify[bot]]

✅ Deploy Preview for localai ready!

Name	Link
🔨 Latest commit	7d199e7f3bf6152b213bc89d5c29939ef26fa76e
🔍 Latest deploy log	https://app.netlify.com/projects/localai/deploys/691175fd7fcc920008755536
😎 Deploy Preview	https://deploy-preview-7211--localai.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.

[mudler]

Description

This PR fixes #7189

Notes for Reviewers

After applying the fix I see the right reranker output:

curl -X POST http://localhost:8080/v1/rerank \
  -H "Content-Type: application/json" \
  -d '{
    "query": "What are the benefits of exercise?",
    "documents": [
      "Regular exercise can improve cardiovascular health.",
      "Eating a balanced diet is important for overall well-being.",
      "Exercise helps in weight management and builds muscle strength.",
      "Reading books can expand your knowledge and vocabulary."
    ],
    "model": "jina-reranker-v1-tiny-en",
    "top_n": 2
  }'


{
  "model": "jina-reranker-v1-tiny-en",
  "usage": {
    "total_tokens": 41,
    "prompt_tokens": 41
  },
  "results": [
    {
      "index": 2,
      "document": {
        "text": "Exercise helps in weight management and builds muscle strength."
      },
      "relevance_score": 0.07032939791679382
    },
    {
      "index": 0,
      "document": {
        "text": "Regular exercise can improve cardiovascular health."
      },
      "relevance_score": 0.059670690447092056
    }
  ]
}

This is great, thanks for opening up a PR!

is there are tests, which I can assert behavior for this backend?

we have tests here that could be likely extended: https://github.com/mudler/LocalAI/blob/bf77c11b6591731c68d6142a68f5eada7c4c5c2d/tests/e2e-aio/e2e_test.go#L286