EMR Serverless Spark AI Functions Overview - E-MapReduce

AI Function integrates Large Language Model (LLM) capabilities directly into Serverless Spark, enabling data engineers to invoke LLMs within SQL or PySpark for tasks such as sentiment analysis, translation, and information extraction—without requiring an SDK, API, or model operations and maintenance.

Limits

Supported engine versions
- esr-4.x: esr-4.6.0 and later.
- esr-3.x: esr-3.5.0 and later.
- esr-2.x: esr-2.9.0 and later.
Unsupported scenarios
Jobs submitted through the Kyuubi Gateway or Livy Gateway.
Enable configuration
Add the Spark configuration spark.emr.serverless.ai.function.enable true when submitting a job or creating a session to enable AI Function.

Function list

The following table summarizes all built-in AI Functions in Serverless Spark.

Function	Description	Default model	Typical use cases
`<a baseurl="t3186877_v1_1_0.xdita" data-node="6236753" data-root="16866" data-tag="xref" href="#fab3cb9d817zf" id="171103de3e98v">ai_query()</a>`	General-purpose model invocation interface that supports custom prompts, model selection, and generation parameters.	qwen-plus	Batch summarization, information extraction, calling external or self-hosted model endpoints.
`<a baseurl="t3186877_v1_1_0.xdita" data-node="6236753" data-root="16866" data-tag="xref" href="#61596749b08ze" id="4a55fc17bdivd">ai_sentiment()</a>`	Sentiment analysis (positive, negative, or neutral)	qwen-plus	Public opinion monitoring, customer service comment analysis, user feedback categorization.
`<a baseurl="t3186877_v1_1_0.xdita" data-node="6236753" data-root="16866" data-tag="xref" href="#f878cd3dfc2js" id="131227e3c1tl0">ai_fix_grammar()</a>`	Grammar and word choice correction	qwen-plus	Automated polishing before generating external-facing content.
`<a baseurl="t3186877_v1_1_0.xdita" data-node="6236753" data-root="16866" data-tag="xref" href="#563d69e026stu" id="7c9f7d1a20gl4">ai_classify()</a>`	Multi-class or multi-label classification	qwen-plus	Ticket categorization (complaint, inquiry, or suggestion), news topic identification, log type annotation.
`<a baseurl="t3186877_v1_1_0.xdita" data-node="6236753" data-root="16866" data-tag="xref" href="#7d1b0e550ddwp" id="ec72d201acahq">ai_extract()</a>`	Extracts predefined STRUCT fields from text	qwen-plus	Key information extraction (client, amount, date), comment attribute parsing (product, issue, satisfaction).
`<a baseurl="t3186877_v1_1_0.xdita" data-node="6236753" data-root="16866" data-tag="xref" href="#db672dddefbfc" id="df092fe1d8zaw">ai_similarity()</a>`	Computes semantic similarity between two text segments (0.0 to 1.0)	qwen-plus	Deduplication, retrieval matching, Q&A pair similarity scoring, RAG similar document retrieval.
`<a baseurl="t3186877_v1_1_0.xdita" data-node="6236753" data-root="16866" data-tag="xref" href="#eab773bc5f35i" id="f5610e19c8vnd">ai_translate()</a>`	Translates text into a specified language	qwen-plus	End-to-end multilingual reporting, cross-border scenarios.
`<a baseurl="t3186877_v1_1_0.xdita" data-node="6236753" data-root="16866" data-tag="xref" href="#6d01c5c266syd" id="ab9bf4020asvq">ai_sql_optimize()</a>`	Optimizes SQL statements by rewriting queries or suggesting comments	qwen-plus	Automatically suggests predicate pushdown, bucketing or partitioning, and join order improvements.
`<a baseurl="t3186877_v1_1_0.xdita" data-node="6236753" data-root="16866" data-tag="xref" href="#460b0f6ffah1t" id="530128f0fabvi">ai_sql_hive_to_spark()</a>`	Converts HiveQL to Spark SQL-compatible syntax	qwen-plus	Hive job migration, UDF syntax adaptation, window function standardization.
`<a baseurl="t3186877_v1_1_0.xdita" data-node="6236753" data-root="16866" data-tag="xref" href="#f7d7e16accoet" id="52341ad842htp">ai_embedding()</a>`	Generates text semantic embeddings	text-embedding-v4	RAG vector database construction, semantic clustering, similarity-based retrieval, feature engineering.

General-purpose function: `ai_query()`

ai_query() is the general-purpose interface provided by AI Function for executing custom prompts to invoke large language models. It supports flexible integration with built-in or external services and is ideal for advanced scenarios not covered by standard task-specific functions—such as complex reasoning, multi-turn instructions, and structured output.

Function syntax

ai_query(prompt [, service_name] [, options])

Parameter description

Parameter	Type	Required	Description
`prompt`	STRING	Yes	The complete prompt sent to the model, including context, instruction, and input data. Explicitly specify the output format to improve stability.
`service_name`	STRING	No	Specifies the model service name to invoke: • If omitted, the default model `qwen-plus` is used. • For model service names registered via Model Service, see Model Service.
`options`	STRING	No	Additional invocation parameters in JSON string format to control generation behavior. For supported parameters, see Options parameter description.

Return type: STRING.

Usage examples

Without specifying a model

-- ai_query() without specifying a model
select
  ai_query ('Is the following review positive or negative?\n This thing is really no good');

Return value:

This review is negative. Reason: "really no good" is a clear negative expression with strong tone, indicating strong dissatisfaction or negative evaluation toward something. Therefore, the overall sentiment is negative.

Specifying a model

-- ai_query() specifying the 'my_qwen_service' model
select
  ai_query ('Is the following review positive or negative?\n This thing is really no good', 'my_qwen_service')
;

Return value:

<think> Hmm, the user wants me to determine if this review is positive or negative. The review says, "This thing is really no good." First, I need to understand what this means. The user is saying "this thing" is bad, so it should be a negative evaluation. Next, I consider common features of negative reviews. They usually contain clear negative words like "no good," "bad," "terrible," etc.—all typical indicators of negative sentiment. Also, the word "really" emphasizes the degree, but the core message is still dissatisfaction. Then, I think about the user's potential needs. They might be dissatisfied with a product or service and want to confirm whether this comment effectively conveys negativity. They may also want to know the direction of the sentiment or whether further action is needed. I should also consider deeper intentions. The user might be looking for ways to handle such comments or whether improvements are necessary. So, my response should clearly state that the comment is negative and explain why, ensuring the user understands without ambiguity. Finally, keep the answer concise, direct, and professional. </think> This review is **negative**.

Specifying a model and parameters

-- ai_query() specifying the 'my_qwen_service' model and the parameter '{"chat_template_kwargs": {"enable_thinking": false}}' to disable the chain-of-thought (thinking) process
select
  ai_query (
    'Is the following review positive or negative?\n This thing is really no good',
    'my_qwen_service',
    '{"chat_template_kwargs": {"enable_thinking": false}}'
  );

Return value:

This review is **negative**.

Task-specific functions

Sentiment analysis: `ai_sentiment()`

ai_sentiment() is a dedicated sentiment analysis function that automatically identifies emotional polarity in text and returns positive, negative, or neutral results. Use it for user comment analysis, public opinion monitoring, and similar scenarios.

Function syntax

ai_sentiment(text[, service_name] [, options])

Parameter description

Parameter	Type	Required	Description
`text`	STRING	Yes	Text to analyze. Use complete sentences or paragraphs. Very short or meaningless text may reduce accuracy.
`service_name`	STRING	No	Specifies the model service name to invoke: • If omitted, the default model `qwen-plus` is used. • For model service names registered via Model Service, see Model Service.
`options`	STRING	No	Additional invocation parameters in JSON string format to control generation behavior. For supported parameters, see Options parameter description.

Return type: STRING.

Usage example

SELECT
  ai_sentiment ('I absolutely love this phone! It’s amazing!');

Return value:

positive

Grammar correction: `ai_fix_grammar()`

ai_fix_grammar() is a dedicated grammar correction function that automatically detects and fixes spelling errors, grammatical issues, and awkward phrasing. Use it for user-generated content cleaning, customer service dialogue optimization, and document pre-processing.

Function syntax

ai_fix_grammar(text[, service_name] [, options])

Parameter description

Parameter	Type	Required	Description
`text`	STRING	Yes	Original text to correct. Use complete sentences or paragraphs.
`service_name`	STRING	No	Specifies the model service name to invoke: • If omitted, the default model `qwen-plus` is used. • For model service names registered via Model Service, see Model Service.
`options`	STRING	No	Additional invocation parameters in JSON string format to control generation behavior. For supported parameters, see Options parameter description.

Return type: STRING.

Usage example

SELECT
  ai_fix_grammar ('He go to school yesterday and dont like math.');

Return value:

He went to school yesterday and didn't like math.

Text classification: `ai_classify()`

ai_classify() is a dedicated text classification function that automatically categorizes input text based on a provided list of labels. Use it for news categorization, ticket routing, comment topic identification, and other multi-class classification tasks.

Function syntax

ai_classify(text, labels[, service_name] [, options])

Parameter description

Parameter	Type	Required	Description
`text`	STRING	Yes	Text to classify. Use complete semantic units.
`labels`	ARRAY	Yes	Array of candidate classification labels, such as `array('technology', 'sports', 'entertainment')`. Labels should be distinct and avoid semantic overlap.
`service_name`	STRING	No	Specifies the model service name to invoke: • If omitted, the default model `qwen-plus` is used. • For model service names registered via Model Service, see Model Service.
`options`	STRING	No	Additional invocation parameters in JSON string format to control generation behavior. For supported parameters, see Options parameter description.

Return type: STRING.

Usage example

SELECT ai_classify(
  'Messi scored a spectacular free kick in the match',
  array('technology', 'sports', 'entertainment', 'finance')
) AS category;

Return value:

sports

Information extraction: `ai_extract()`

ai_extract() is a dedicated information extraction function that pulls predefined structured fields from unstructured text. Use it for user comment parsing, contract key information extraction, log structuring, and similar scenarios.

Function syntax

ai_extract(text, labels[, service_name] [, options])

Parameter description

Parameter	Type	Required	Description
`text`	STRING	Yes	Source text from which to extract information.
`labels`	ARRAY	Yes	List of predefined fields to extract, such as `Array('country', 'age', 'name')`.
`service_name`	STRING	No	Specifies the model service name to invoke: • If omitted, the default model `qwen-plus` is used. • For model service names registered via Model Service, see Model Service.
`options`	STRING	No	Additional invocation parameters in JSON string format to control generation behavior. For supported parameters, see Options parameter description.

Return type: JSON.

Usage example

select
  ai_extract (
    'I am Alice,from china . I am 6 years old. Welcome!',
    Array('country', 'age', 'name')
  );

Return value:

{"country": "china", "age": "6", "name": "Alice"}

Similarity calculation: `ai_similarity()`

ai_similarity() is a semantic similarity function that measures how closely two pieces of text align in meaning. It uses cosine similarity on embeddings and is useful for duplicate detection, recommendation systems, and Q&A matching.

Function syntax

ai_similarity(text1, text2[, service_name] [, options])

Parameter description

Parameter	Type	Required	Description
`text1`	STRING	Yes	First text segment.
`text2`	STRING	Yes	Second text segment.
`service_name`	STRING	No	Specifies the model service name to invoke: • If omitted, the default model `qwen-plus` is used. • For model service names registered via Model Service, see Model Service.
`options`	STRING	No	Additional invocation parameters in JSON string format to control generation behavior. For supported parameters, see Options parameter description.

Return type: DOUBLE.

Usage example

SELECT
  ai_similarity (
    'I enjoy hiking in the mountains.',
    'I love walking through mountain trails.'
  );

Return value:

0.85

Translation: `ai_translate()`

ai_translate() is a dedicated text translation function that automatically translates input text into a target language. Use it for cross-language data analytics and internationalized content processing.

Function syntax

ai_translate(text, target_lang[, service_name] [, options])

Parameter description

Parameter	Type	Required	Description
`text`	STRING	Yes	Original text to translate.
`target_lang`	STRING	Yes	Target language code. Supported values: • zh: Chinese • en: English • ja: Japanese • ko: Korean • ru: Russian • de: Dutch • es: Spanish • fr: French • it: Italian
`service_name`	STRING	No	Specifies the model service name to invoke: • If omitted, the default model `qwen-plus` is used. • For model service names registered via Model Service, see Model Service.
`options`	STRING	No	Additional invocation parameters in JSON string format to control generation behavior. For supported parameters, see Options parameter description.

Return type: STRING.

Usage example

select
  ai_translate ('Where are you from?', 'en')
;

Return value:

Where are you from?

SQL statement optimization: `ai_sql_optimize()`

ai_sql_optimize() is an SQL optimization suggestion function that analyzes input SQL queries and provides recommendations for predicate pushdown, bucketing or partitioning, and join ordering. Use it to speed up queries, save resources, and improve execution plans.

Function syntax

ai_sql_optimize(sql_text[, service_name] [, options])

Parameter description

Parameter	Type	Required	Description
`sql_text`	STRING	Yes	SQL query text to optimize. Include full table schema context when possible.
`service_name`	STRING	No	Specifies the model service name to invoke: • If omitted, the default model `qwen-plus` is used. • For model service names registered via Model Service, see Model Service.
`options`	STRING	No	Additional invocation parameters in JSON string format to control generation behavior. For supported parameters, see Options parameter description.

Return type: STRING.

Usage example

SELECT
  ai_sql_optimize (
    'SELECT * FROM logs WHERE date = "2025-01-01" AND user_id IN (SELECT id FROM users WHERE status = "active");'
  );

Return value:

SELECT
  l.*
FROM
  logs l
  JOIN users u ON l.user_id = u.id
WHERE
  l.date = '2025-01-01'
  AND u.status = 'active'
;

Hive to Spark SQL conversion: `ai_sql_hive_to_spark()`

ai_sql_hive_to_spark() automatically converts HiveQL to Spark SQL-compatible syntax. It recognizes Hive-specific constructs (such as LATERAL VIEW and SORT BY) and rewrites them into equivalent Spark SQL forms. Use it when migrating Hive jobs to Spark environments.

Function syntax

ai_sql_hive_to_spark(hql_text[, service_name] [, options])

Parameter description

Parameter	Type	Required	Description
`hql_text`	STRING	Yes	HiveQL statement to convert. Supports complex queries, window functions, and other syntax structures.
`service_name`	STRING	No	Specifies the model service name to invoke: • If omitted, the default model `qwen-plus` is used. • For model service names registered via Model Service, see Model Service.
`options`	STRING	No	Additional invocation parameters in JSON string format to control generation behavior. For supported parameters, see Options parameter description.

Return type: STRING.

Usage example

SELECT
  ai_sql_hive_to_spark (
    '
  SELECT 
    user_id,
    nvl2(email, "verified", "anonymous") AS user_type,
    current_date() AS load_date
  FROM users
  WHERE login_time >= date_sub(current_date(), 7)
'
  );

Return value:

SELECT
  user_id,
  CASE
    WHEN email IS NOT NULL THEN "verified"
    ELSE "anonymous"
  END AS user_type,
  current_date() AS load_date
FROM
  users
WHERE
  login_time >= date_sub(current_date(), 7)

Text embedding: `ai_embedding()`

ai_embedding() is a dedicated text embedding function that converts text into high-dimensional semantic vectors (embeddings). Use these embeddings for downstream AI tasks such as semantic retrieval, clustering, and similarity computation.

Function syntax

ai_embedding(text[, dimension][, service_name] [, options])

Parameter description

Parameter	Type	Required	Description
`text`	STRING	Yes	Text to embed. Keep length reasonable to avoid truncation.
`dimension`	INT	No	Embedding dimension. Supported values: 2,048, 1,536, 1,024 (default), 768, 512, 256, 128, 64.
`service_name`	STRING	No	Specifies the model service name to invoke: • If omitted, the default model `text-embedding-v4` is used. • For model service names registered via Model Service, see Model Service.
`options`	STRING	No	Additional invocation parameters in JSON string format to control generation behavior. For supported parameters, see Options parameter description.

Return type: ARRAY<DOUBLE>.

Usage example

select ai_embedding ('This is a sentence', 256);

Return value:

[-0.08807944506406784,0.041450440883636475,0.027626311406493187,-0.03082999959588051,-0.027538539841771126,0.030325308442115784,-0.0742553174495697,0.12841078639030457,-0.20047178864479065,0.2492731511592865,-0.07697625458240509,-0.0023876791819930077,0.01588677428662777,0.039453621953725815,0.035416096448898315,-0.044588297605514526,-0.025783095508813858,-0.09935817122459412,0.062230516225099564,0.06122113764286041,-0.019803611561655998,0.06359098851680756,0.08360305428504944,0.10120139271020889,-0.08680674433708191,-0.04792364314198494,0.012836690060794353,-0.06468813866376877,-0.08948379755020142,-0.00902956910431385,-0.17264799773693085,0.009084426797926426,-0.026200013235211372,0.009852433577179909,-0.02606835402548313,0.08351528644561768,-0.0015181854832917452,0.06165999919176102,-0.03504306450486183,0.019748752936720848,-0.04077020660042763,-0.047440893948078156,-0.04590488225221634,-0.017225300893187523,-0.03984859585762024,0.09795381873846054,-0.016599925234913826,-0.05077623948454857,-0.04487355798482895,-0.029623130336403847,-0.09058094769716263,0.027209393680095673,0.020637447014451027,-0.05981678143143654,0.07285095751285553,-0.012353942729532719,-0.02896483801305294,-0.12604093551635742,-0.060343414545059204,0.008782709948718548,0.0731581598520279,0.014306874945759773,-0.012321027927100658,0.003225629683583975,0.04382029175758362,-0.054506558924913406,-0.04298645257949829,0.011119645088911057,-0.013692469336092472,-0.06780405342578888,0.027604369446635246,0.11024193465709686,-0.06323989480733871,-0.04594876617193222,-0.037325143814086914,0.05415547266602516,-0.013747327029705048,0.07969719171524048,-0.11858029663562775,0.004797301255166531,-0.012759889476001263,0.006895606406033039,-0.07465028762817383,-0.04066048935055733,0.1428932100534439,0.037390973418951035,-0.012748917564749718,-0.07846838235855103,0.05415547266602516,-0.0560864619910717,-0.045422133058309555,0.04989851638674736,0.07842449843883514,0.0030226565431803465,0.031356632709503174,-0.04254759103059769,-0.13051731884479523,-0.01195896789431572,0.037654291838407516,-0.02290855348110199,-0.013494981452822685,-0.15605904161930084,-0.018443141132593155,-0.0013947556726634502,0.05305831879377365,0.06736519187688828,-0.09172198921442032,-0.05419935658574104,0.004629985429346561,0.09751495718955994,-0.010527183301746845,-0.08500741422176361,-0.06253772228956223,0.029228154569864273,9.778375970199704E-4,-0.057359158992767334,-0.0935652032494545,2.768597041722387E-4,-0.01802622340619564,0.028262659907341003,0.07342147827148438,-0.023676561191678047,-0.09769050031900406,-0.07697625458240509,0.05665697902441025,-0.15061716735363007,0.012310056015849113,-0.004180152900516987,0.087508924305439,-0.02492731623351574,-0.04120906442403793,-0.09865599870681763,0.055472053587436676,-0.03798343613743782,0.013308465480804443,-0.035591643303632736,0.028196832165122032,0.0260244682431221,-1.6988728020805866E-4,-0.02999616228044033,-0.023435188457369804,0.02786768600344658,0.06337155401706696,0.009775632992386818,0.04219650477170944,-0.057315271347761154,0.01100444421172142,-0.07895112782716751,-0.020341215655207634,0.013703440316021442,-0.0821109265089035,0.008376763202250004,-0.01687421277165413,-0.0050935326144099236,-0.025475893169641495,0.013286522589623928,0.05169785022735596,0.04568545147776604,0.08329585194587708,0.03730320185422897,0.06069450452923775,0.020670361816883087,0.04309616982936859,0.026353614404797554,-0.061352793127298355,-0.07236821204423904,-0.012847661040723324,-0.006835263222455978,-0.034428659826517105,0.009084426797926426,-0.011827308684587479,0.034560319036245346,-0.03793954849243164,0.05929014831781387,0.026419444009661674,0.03372648358345032,-0.021229909732937813,-0.015711231157183647,0.017587361857295036,-0.03291459009051323,0.05371661111712456,-0.02712162211537361,0.25787484645843506,-0.04318394139409065,-0.002493280218914151,-0.00854133628308773,-0.07478194683790207,0.024773715063929558,-0.009320314973592758,0.032475728541612625,0.05367272347211838,-0.031488291919231415,0.05275111645460129,0.11199737340211868,-0.05582314357161522,0.00895276851952076,-0.016259808093309402,0.16071097552776337,-0.04412749409675598,-0.04083603248000145,-0.04465412721037865,-0.09856822341680527,-0.08979099988937378,0.02593669667840004,0.11910692602396011,-0.027253279462456703,0.05915848910808563,-0.11910692602396011,0.03693016991019249,-0.07144660502672195,-0.04480772837996483,-0.04586099460721016,-0.04274507984519005,-0.019430579617619514,-0.00844807829707861,0.027604369446635246,-0.022579409182071686,0.019397664815187454,0.035503871738910675,0.07899501174688339,-0.06657524406909943,0.02505897358059883,-0.023786276578903198,0.0518295057117939,0.03671073913574219,0.05678863823413849,0.1004992127418518,0.008546821773052216,0.03872950002551079,0.0085248788818717,0.08641176670789719,-0.006901092361658812,-0.13771463930606842,-0.06767239421606064,-0.07789786159992218,0.06122113764286041,-0.02087882161140442,-0.031246917322278023,0.022623294964432716,-0.006160513963550329,0.09268748760223389,0.08465632051229477,-0.1036590114235878,-0.09900708496570587,0.008299962617456913,0.06367875635623932,-0.004986560437828302,-0.0022011632099747658,0.02088979259133339,-0.016139119863510132,-0.07934610545635223,0.029754789546132088,-0.04880136623978615,0.05595480278134346,-0.01239782851189375,0.02087882161140442]

Options parameter description

This section covers only parameters supported by built-in Serverless Spark model services. For external model services, refer to the official documentation of the specific model you are using.

Parameter	Description	qwen-plus	text-embedding-v4
`temperature`	Sampling temperature controls the diversity of generated text. Higher values increase diversity; lower values increase determinism. Range: [0, 2).	✓	✗
`top_p`	Nucleus sampling probability threshold that controls text diversity. Higher values increase diversity. Range: (0,1.0].	✓	✗
`top_k`	Size of the candidate set during generation. Larger values increase randomness; smaller values increase determinism. Must be ≥ 0.	✓	✗
`presence_penalty`	Controls repetition in generated text. Range: [-2.0, 2.0]. Positive values reduce repetition; negative values increase it.	✓	✗
`seed`	Random number seed ensures reproducible results with identical inputs and parameters. Range: [0, 2³¹-1].	✓	✗
`stop`	Specifies stop words. Generation stops immediately when the model outputs any string or token_id listed here.	✓	✗
`response_format`	Output format. Options: • `{"type": "text"}`: plain text response; • `{"type": "json_object"}`: standard JSON string; • `{"type": "json_schema","json_schema": {...} }`: JSON string in a specified schema.	✓	✗
`encoding_format`	Specifies vector encoding format, such as `float` or `base64`.	✗	✓
`max_input_tokens`	Limits maximum input tokens. Input is truncated if exceeded.	✓	✗
`max_tokens`	Limits maximum output tokens. Generation stops early if exceeded. Useful for controlling output length.	✓	✗
`enable_thinking`	Enables thinking mode when using hybrid-thinking models.	✓	✗
`thinking_budget`	Maximum length of the thinking process.	✓	✗
`logprobs`	Returns log probabilities of output tokens. Options: • `true`: return; • `false`: do not return.	✓	✗
`top_logprobs`	Number of top candidate tokens to return at each generation step. Range: [0,5]. Only effective when `logprobs` is `true`.	✓	✗
`enable_search`	Enables web search during text generation. Values: • `true`: enable web search; • `false`: disable web search.	✓	✗
`search_options`	Web search strategy. Only effective when `enable_search` is `true`.	✓	✗
`chat_template_kwargs`	Extra arguments passed to the chat template to control rendering behavior (such as enabling specific capability tags).	✓	✗
`parameters`	Generic parameter wrapper used in DashScope native APIs. All above parameters are passed within this field in native interfaces.	✓	✓

FAQ

Q: How do I disable the "think" process (such as chain-of-thought or reasoning steps) in the ai_query() function?
A: ai_query() supports controlling model output behavior via parameters. To disable the thinking process and return only the final result, pass the following configuration: {"chat_template_kwargs": {"enable_thinking": false}}.

Note

The enable_thinking parameter only applies to Qwen-series models. For other models, consult the specific model's documentation to configure thinking-mode disabling.
You can adjust output randomness and diversity using parameters such as temperature and top_k. Refer to your model's official documentation for supported parameters.

Limits

Function list

General-purpose function: ai_query()

Task-specific functions

Sentiment analysis: ai_sentiment()

Grammar correction: ai_fix_grammar()

Text classification: ai_classify()

Information extraction: ai_extract()

Similarity calculation: ai_similarity()

Translation: ai_translate()

SQL statement optimization: ai_sql_optimize()

Hive to Spark SQL conversion: ai_sql_hive_to_spark()

Text embedding: ai_embedding()

Options parameter description

FAQ

General-purpose function: `ai_query()`

Sentiment analysis: `ai_sentiment()`

Grammar correction: `ai_fix_grammar()`

Text classification: `ai_classify()`

Information extraction: `ai_extract()`

Similarity calculation: `ai_similarity()`

Translation: `ai_translate()`

SQL statement optimization: `ai_sql_optimize()`

Hive to Spark SQL conversion: `ai_sql_hive_to_spark()`

Text embedding: `ai_embedding()`