BAAI/bge-reranker-v2-m3

BAAI/bge-reranker-v2-m3 is a lightweight reranker model with strong multilingual capabilities. It’s easy to deploy and offers fast inference.

Model details

Category	Details
Model Name	BAAI/bge-reranker-v2-m3
Version	Original
Model Category	Reranker
Size	568M parameters
HuggingFace Model	BAAI/bge-reranker-v2-m3
License	Apache 2.0

Capabilities

Feature	Details
Azion Long-term Support (LTS)	❌
Context Length	8k tokens
Supports LoRA	❌
Input data	Text

Usage

Rerank example

This is an example of a basic rerank request using this model:

const modelResponse = await Azion.AI.run("baai-bge-reranker-v2-m3", {
  "query": "What is deep learning?",
  "documents": [
    "Deep learning is a subset of machine learning that uses neural networks with many layers",
    "The weather is nice today",
    "Deep learning enables computers to learn from large amounts of data",
    "I like pizza"
  ]
})

Property	Type	Description
`query`	string	The search query or prompt to rank the documents against.
`documents`	string[]	An array of documents to be ranked based on their relevance to the query.

Score example

This is an example of a basic score request using this model:

const modelResponse = await Azion.AI.run("baai-bge-reranker-v2-m3", {
  "text_1": "What is deep learning?",
  "text_2": [
    "Deep learning is a subset of machine learning that uses neural networks with many layers",
    "The weather is nice today",
    "Deep learning enables computers to learn from large amounts of data",
    "I like pizza"
  ]
})

Property	Type	Description
`text_1`	string	The first text input for the model to process.
`text_2`	string[]	An array of text inputs for the model to process and give a score.

Response example:

{
  "id": "rerank-356bf11f0e794f3c8f726bec7ba698bb",
  "model": "baai-bge-reranker-v2-m3",
  "usage": {
    "total_tokens": 78
  },
  "results": [
    {
      "index": 0,
      "document": {
        "text": "Deep learning is a subset of machine learning that uses neural networks with many layers"
      },
      "relevance_score": 0.99951171875
    },
    {
      "index": 2,
      "document": {
        "text": "Deep learning enables computers to learn from large amounts of data"
      },
      "relevance_score": 0.98291015625
    },
    {
      "index": 3,
      "document": {
        "text": "I like pizza"
      },
      "relevance_score": 0.00001621246337890625
    },
    {
      "index": 1,
      "document": {
        "text": "The weather is nice today"
      },
      "relevance_score": 0.000016033649444580078
    }
  ]
}

Property	Type	Description
`id`	string	Unique identifier for the rerank request.
`model`	string	The name of the model used for reranking.
`usage.total_tokens`	number	The total number of tokens used in the request.
`results[]`	object[]	An array of reranked result objects.
`results[].index`	number	The index of the document in the input list.
`results[].document`	object	The document object containing the text.
`results[].document.text`	string	The textual content of the document.
`results[].relevance_score`	number	The relevance score assigned to the document by the model.

JSON schema

{
    "$schema": "http://json-schema.org/draft-07/schema#",
    "type": "object",
    "required": [
        "query",
        "documents"
    ],
    "properties": {
        "query": {
            "type": "string"
        },
        "documents": {
            "type": "array",
            "items": {
                "type": "string"
            }
        },
        "top_n": {
            "type": "integer"
        },
        "max_tokens_per_doc": {
            "type": "integer"
        }
    }
}