AI_SIMILARITY

Syntax

FLOAT AI_SIMILARITY(
  STRING <model_name>,
  STRING <version_name>,
  STRING <input1>,
  STRING <input2>
  [, STRING <model_parameters>]
);

Parameters

• model_name: Required. A STRING. The name of the model to use. For more information, see SQL AI functions.

• version_name: Required. A STRING. The name of the model version to use. To call the default version, specify DEFAULT_VERSION.

• input1: Required. A STRING. The first text for the semantic similarity comparison.

• input2: Required. A STRING. The second text for the semantic similarity comparison.

• model_parameters: Optional. STRING. Specifies model parameters such as max_tokens, temperature, and top_p. The format is a JSON string:

  '{"max_tokens": 500, "temperature": 0.6, "top_p": 0.95}'.

  • max_tokens: The maximum number of tokens to generate in a single model call. For MaxCompute public models, the default value is 4,096.

  • temperature: A value between 0 and 1 that controls the randomness of the output. A higher value results in more creative and diverse output, while a lower value produces more deterministic and conservative output.

  • top_p: A value between 0 and 1 that limits the range of candidate labels the model can choose from. A higher value allows for a broader range and more diversity, while a lower value narrows the range and produces more focused results.

Return value

Returns a FLOAT value between 0 and 1 that represents the similarity score of the two input texts. The return values are described as follows:

• A score of 1.0 indicates that the two texts are identical. A score of 0.0 indicates that they are completely unrelated.

• If input1 or input2 is NULL, the function returns NULL.

• If input1 or input2 is not a STRING, the function returns an error.

Examples

SET odps.namespace.schema=true;

SELECT AI_SIMILARITY(
    bigdata_public_modelset.default.`qwen3.7-max`,
    DEFAULT_VERSION,
    'MaxCompute is a big data computing platform.',
    'MaxCompute provides large-scale data processing capabilities.'
) AS similarity_score;
-- Result
+------------------+
| similarity_score |
+------------------+
| 0.75             |
+------------------+

-- Sample data
CREATE TABLE text_pairs (
    text1 STRING,
    text2 STRING
);

INSERT INTO text_pairs VALUES
    ('Cloud computing supports scalable infrastructure.', 'Enterprises can flexibly scale IT resources using cloud services.'),
    ('The weather is sunny today.', 'Machine learning algorithms require large datasets.'),
    ('A data warehouse stores historical data for analysis.', 'A data warehouse provides analytical processing capabilities for large-scale data.'),
    ('I like to read books.', 'Reading is my favorite hobby.');

-- Compare the semantic similarity of multiple text pairs
SET odps.namespace.schema=true;

SELECT
    text1,
    text2,
    AI_SIMILARITY(
        bigdata_public_modelset.default.`deepseek-v4-pro`,
        DEFAULT_VERSION,
        text1,
        text2
    ) AS similarity_score
FROM text_pairs
ORDER BY similarity_score DESC;

-- Result:
+----------------------------------------------------+--------------------------------------------------------------------------+------------------+
| text1                                              | text2                                                                    | similarity_score |
+----------------------------------------------------+--------------------------------------------------------------------------+------------------+
| A data warehouse stores historical data for analysis. | A data warehouse provides analytical processing capabilities for large-scale data. | 0.92             |
| I like to read books.                              | Reading is my favorite hobby.                                            | 0.88             |
| Cloud computing supports scalable infrastructure.  | Enterprises can flexibly scale IT resources using cloud services.        | 0.82             |
| The weather is sunny today.                        | Machine learning algorithms require large datasets.                      | 0.05             |
+----------------------------------------------------+--------------------------------------------------------------------------+------------------+

FAILED: ODPS-0130071:[1,8] Semantic analysis exception - inference quota status check failed, error message: region cn-shanghai not found in inference quota