Large Language Model (LLM) Functions (Snowflake Cortex)

Snowflake Cortex gives you instant access to industry-leading large language models (LLMs) trained by researchers at companies like Mistral, Reka, Meta, and Google, including Snowflake Arctic, an open enterprise-grade model developed by Snowflake.

Since these LLMs are fully hosted and managed by Snowflake, using them requires no setup. Your data stays within Snowflake, giving you the performance, scalability, and governance you expect.

Snowflake Cortex features are provided as SQL functions and are also available in Python. The available functions are summarized below.

  • COMPLETE: Given a prompt, returns a response that completes the prompt. This function accepts either a single prompt or a conversation with multiple prompts and responses.

  • EMBED_TEXT_768: Given a piece of text, returns a vector embedding of 768 dimensions that represents that text.

  • EMBED_TEXT_1024: Given a piece of text, returns a vector embedding of 1024 dimensions that represents that text.

  • EXTRACT_ANSWER: Given a question and unstructured data, returns the answer to the question if it can be found in the data.

  • SENTIMENT: Returns a sentiment score, from -1 to 1, representing the detected positive or negative sentiment of the given text.

  • SUMMARIZE: Returns a summary of the given text.

  • TRANSLATE: Translates given text from any supported language to any other.

The following helper functions are available as SQL functions:

  • COUNT_TOKENS: Given an input text, returns the token count based on the model or Cortex function specified.

  • TRY_COMPLETE: Operates the same as the COMPLETE function, except that it returns NULL when the function could not execute instead of an error code.

Required privileges

The CORTEX_USER database role in the SNOWFLAKE database includes the privileges that allow users to call Snowflake Cortex LLM functions. By default, the CORTEX_USER role is granted to the PUBLIC role. The PUBLIC role is automatically granted to all users and roles, so this allows all users in your account to use the Snowflake Cortex LLM functions.

If you don’t want all users to have this privilege, you can revoke access to the PUBLIC role and grant access to specific roles.

To revoke the CORTEX_USER database role from the PUBLIC role, run the following command using the ACCOUNTADMIN role:

REVOKE DATABASE ROLE SNOWFLAKE.CORTEX_USER
  FROM ROLE PUBLIC;
Copy

You can then selectively provide access to specific roles. The SNOWFLAKE.CORTEX_USER database role cannot be granted directly to a user. For more information, see Using SNOWFLAKE database roles. A user with the ACCOUNTADMIN role can grant this role to a custom role in order to allow users to access Cortex LLM Functions. In the following example, use the ACCOUNTADMIN role and grant the user some_user the CORTEX_USER database role via the account role cortex_user_role, which you create for this purpose.

USE ROLE ACCOUNTADMIN;

CREATE ROLE cortex_user_role;
GRANT DATABASE ROLE SNOWFLAKE.CORTEX_USER TO ROLE cortex_user_role;

GRANT ROLE cortex_user_role TO USER some_user;
Copy

You can also grant access to Snowflake Cortex LLM functions through existing roles commonly used by specific groups of users. (See User roles.) For example, if you have created an analyst role that is used as a default role by analysts in your organization, you can easily grant these users access to Snowflake Cortex LLM functions with a single GRANT statement.

GRANT DATABASE ROLE SNOWFLAKE.CORTEX_USER TO ROLE analyst;
Copy

Availability

Snowflake Cortex LLM functions are currently available in the following regions.

Function
(Model)
AWS US West 2
(Oregon)
AWS US East 1
(N. Virginia)
AWS Europe Central 1
(Frankfurt)
AWS AP Southwest 2
(Sydney)
Azure East US 2
(Virginia)
Azure West Europe
(Netherlands)
COMPLETE
(llama3-8b)

COMPLETE
(llama3-70b)

COMPLETE
(snowflake-arctic)

COMPLETE
(reka-core)

COMPLETE
(reka-flash)

COMPLETE
(mistral-large)

COMPLETE
(mixtral-8x7b)

COMPLETE
(llama2-70b-chat)

COMPLETE
(mistral-7b)

COMPLETE
(gemma-7b)

EMBED_TEXT_768
(e5-base-v2)

EMBED_TEXT_768
(snowflake-arctic-embed-m)

EMBED_TEXT_1024
(nv-embed-qa-4)

EXTRACT_ANSWER

SENTIMENT

SUMMARIZE

TRANSLATE

Note

  • The COUNT_TOKENS function is available in all regions, but model inference is region-specific, as per the table.

  • TRY_COMPLETE operate across the same regions as the COMPLETE function.

Cost considerations

Snowflake Cortex LLM functions incur compute cost based on the number of tokens processed. Refer to the consumption table for each function’s cost in credits per million tokens.

Note

Usage of Snowflake Arctic does not incur compute cost through August 1, 2024.

A token is the smallest unit of text processed by Snowflake Cortex LLM functions, approximately equal to four characters of text. The equivalence of raw input or output text to tokens can vary by model.

  • For functions that generate new text in the response (COMPLETE, SUMMARIZE, and TRANSLATE), both input and output tokens are counted.

  • For the EMBED_TEXT_* functions, only input tokens are counted.

  • For functions that only extract information from the input (EXTRACT_ANSWER and SENTIMENT), only input tokens are counted.

  • For EXTRACT_ANSWER, the number of billable tokens is the sum of the number of tokens in the from_text and question fields.

  • SUMMARIZE, TRANSLATE, EXTRACT_ANSWER, and SENTIMENT add a prompt to the input text in order to generate the response. As a result, the input token count is slightly higher than the number of tokens in the text you provide.

  • TRY_COMPLETE does not incur costs for error handling. This means that if the TRY_COMPLETE function returns NULL, no cost is incurred.

  • COUNT_TOKENS incurs only compute cost to run the function. No additional token based costs are incurred.

Snowflake recommends executing queries that call a Snowflake Cortex LLM Function with a smaller warehouse (no larger than MEDIUM) because larger warehouses do not increase performance. The cost associated with keeping a warehouse active will continue to apply when executing a query that calls a Snowflake Cortex LLM Function. For general information on compute costs, see Understanding compute cost.

Track costs for AI services

To track credits used for AI Services including LLM Functions in your account, use the METERING_HISTORY view:

SELECT *
  FROM snowflake.account_usage.metering_daily_history
  WHERE SERVICE_TYPE='AI_SERVICES'
Copy

Usage quotas

To ensure that all Snowflake customers can access LLM capabilities, Snowflake Cortex LLM functions may be subject to throttling during periods of high utilization. Usage quotas are not applied at the account level.

Throttled requests will receive an error response and should be retried later.

Note

On-demand Snowflake accounts without a valid payment method (such as trial accounts) are limited to roughly one credit per day in Snowflake Cortex LLM function usage. To remove this restriction, convert your trial account to a paid account.

Managing costs and throttling

Snowflake recommends using a warehouse size no larger than MEDIUM when calling Snowflake Cortex LLM functions. Using a larger warehouse than necessary does not increase performance, but can result in unnecessary costs and a higher risk of throttling. This recommendation may not apply in the future due to upcoming product updates.

Model restrictions

Models used by Snowflake Cortex have limitations on size as described in the table below. Sizes are given in tokens. Tokens generally correspond to words, but not all tokens are words, so the number of words corresponding to a limit is slightly less than the number of tokens. Inputs that exceed the limit result in an error.

Function

Model

Context window (tokens)

COMPLETE

snowflake-arctic

4,096

mistral-large

32,000

reka-flash

100,000

reka-core

32,000

mixtral-8x7b

32,000

llama2-70b-chat

4,096

llama3-8b

8,000

llama3-70b

8,000

mistral-7b

32,000

gemma-7b

8,000

EMBED_TEXT_768

e5-base-v2

512

snowflake-arctic-embed-m

512

EMBED_TEXT_1024

nv-embed-qa-4

512

EXTRACT_ANSWER

Snowflake managed model

2,048 for text
64 for question

SENTIMENT

Snowflake managed model

512

SUMMARIZE

Snowflake managed model

32,000

TRANSLATE

Snowflake managed model

1,024

Choosing a model

The Snowflake Cortex COMPLETE function supports multiple models of varying capability, latency, and cost. These models have been carefully chosen to align with common customer use cases. To achieve the best performance per credit, choose a model that’s a good match for the content size and complexity of your task. Here are brief overviews of the available models.

Large models

If you’re not sure where to start, try the most capable models first to establish a baseline to evaluate other models. reka-core and mistral-large are the most capable models offered by Snowflake Cortex, and will give you a good idea what a state-of-the-art model can do.

  • reka-core is Reka AI’s most advanced large language model with strong reasoning abilities, code generation, and multilingual fluency.

  • mistral-large is Mistral AI’s most advanced large language model with top-tier reasoning capabilities. Mistral Large is ideal for complex tasks that require large reasoning capabilities or are highly specialized, such as synthetic text generation, code generation, or agents.

Medium models

  • snowflake-arctic is Snowflake’s top-tier enterprise-focused LLM. Arctic excels at enterprise tasks such as SQL generation, coding and instruction following benchmarks.

  • reka-flash is a highly capable multilingual language model optimized for fast workloads that require high quality, such as writing product descriptions or blog posts, coding, and extracting answers from documents with hundreds of pages.

  • mixtral-8x7b is ideal for text generation, classification, and question answering. Mistral models are optimized for low latency with low memory requirements, which translates into higher throughput for enterprise use cases.

  • llama3-70b is an open source model that demonstrates state-of-the-art performance ideal for chat applications, content creation, and enterprise applications.

  • llama2-70b-chat is well-suited to tasks that require a low to moderate amount of reasoning, like extracting data or helping you to write job descriptions.

Small models

  • llama3-8b is ideal for tasks that require low to moderate reasoning with better accuracy than the llama2-70b-chat, like text classification, summarization, and sentiment analysis.

  • mistral-7b is ideal for your simplest summarization, structuration, and question answering tasks that need to be done quickly. It offers low latency and high throughput processing for multiple pages of text with its 32K context window.

  • gemma-7b is suitable for simple code and text completion tasks. It has a context window of 8,000 tokens but is surprisingly capable within that limit, and quite cost-effective.

The following table provides information on how popular models perform on various benchmarks, including the models offered by Snowflake Cortex COMPLETE as well as a few other popular models.

Model

Context Window
(Tokens)
MMLU
(Reasoning)
MT-Bench
(Instruction Following)
HumanEval
(Coding)
Spider 1.0
(SQL)

Snowflake Arctic

4,096

67.3

-

64.3

79.0

GPT 4 Turbo*

32,000

86.4

8.96

67

86.6

reka-core

32,000

83.2

-

76.8

-

llama3-70b

8,000

82

-

81.7

80.2

mistral-large

32,000

81.2

-

45.1

81

Claude 2*

100,000

78.5

8.06

71.2

-

reka-flash

100,000

75.9

8.2

72

-

mixtral-8x7b

32,000

70.6

8.30

40.2

-

GPT 3.5 Turbo*

4,097

70

8.39

48.1

-

llama-2-70b-chat

4,096

68.9

6.86

30.5

-

llama3-8b

8,000

68.4

-

62.2

69.9

mistral-7b

32,000

62.5

6.84

26.2

-

gemma-7b

8,000

64.3

-

32.3

-

llama2-7b*

4,096

45.3

6.27

12.2

-

*Provided for comparison; not available in Snowflake Cortex COMPLETE.

LLM functions overview

Cortex LLM Functions can be grouped into the following categories:

  • COMPLETE function: General purpose function that can perform a wide range of user-specified tasks, such as aspect-based sentiment classification, synthetic data generation, and customized summaries. You can also use this function with your fine-tuned models.

  • Task-specific functions: Purpose-built and managed functions that automate routine-tasks, like simple summaries and quick translations, that don’t require any customization.

  • Helper functions: Purpose-built and managed functions that reduce cases of failures when running other LLM functions, for example by getting the count of tokens in an input prompt to ensure the call doesn’t exceed a model limit.

COMPLETE

Given a prompt, the instruction-following COMPLETE function generates a response using your choice of language model. In the simplest use case, the prompt is a single string. You may also provide a conversation including multiple prompts and responses for interactive chat-style usage, and in this form of the function you can also specify hyperparameter options to customize the style and size of the output.

The COMPLETE function supports the following models. Different models can have different costs.

  • snowflake-arctic

  • mistral-large

  • reka-flash

  • reka-core

  • mixtral-8x7b

  • llama2-70b-chat

  • llama3-8b

  • llama3-70b

  • mistral-7b

  • gemma-7b

See COMPLETE (SNOWFLAKE.CORTEX) for syntax and examples.

Task-specific functions

Given some input text, task-specific functions execute the task for which it was designed without needing to specify a prompt. Task-specific functions quickly and cost-effectively execute routine tasks that don’t require any customization.

EMBED_TEXT_768

The EMBED_TEXT_768 function creates a vector embedding of 768 dimensions for a given English-language text. To learn more about embeddings and vector comparison functions, see Vector Embeddings.

For syntax and examples, see EMBED_TEXT_768 (SNOWFLAKE.CORTEX).

EMBED_TEXT_1024

The EMBED_TEXT_1024 function creates a vector embedding of 1024 dimensions for a given English-language text. To learn more about embeddings and vector comparison functions, see Vector Embeddings.

For syntax and examples, see EMBED_TEXT_1024 (SNOWFLAKE.CORTEX).

EXTRACT_ANSWER

The EXTRACT_ANSWER function extracts an answer to a given question from a text document. The document may be a plain-English document or a string representation of a semi-structured (JSON) data object.

For syntax and examples, see EXTRACT_ANSWER (SNOWFLAKE.CORTEX).

SENTIMENT

The SENTIMENT function returns sentiment as a score between -1 to 1 (with -1 being the most negative and 1 the most positive, with values around 0 neutral) for the given English-language input text.

For syntax and examples, see SENTIMENT (SNOWFLAKE.CORTEX).

SUMMARIZE

The SUMMARIZE function returns a summary of the given English text.

For syntax and examples, see SUMMARIZE (SNOWFLAKE.CORTEX).

TRANSLATE

The TRANSLATE function translates text from the indicated or detected source language to a target language.

For syntax and examples, see TRANSLATE (SNOWFLAKE.CORTEX).

Helper functions

Helper functions are managed functions that are built to help reduce errors when running other Cortex LLM functions.

COUNT_TOKENS

The COUNT_TOKENS function calculates the number of tokens in a prompt for the large language model specified in COMPLETE, and the input text for task-specific functions.

For syntax and examples, see COUNT_TOKENS (SNOWFLAKE.CORTEX).

TRY_COMPLETE

The TRY_COMPLETE function performs the same operation as the COMPLETE function but returns NULL instead of raising an error when the operation cannot be performed.

For syntax and examples, see TRY_COMPLETE (SNOWFLAKE.CORTEX).

Error conditions

Snowflake Cortex LLM functions can produce the following error messages.

Message

Explanation

too many requests

The request was rejected due to excessive system load. Please try your request again.

invalid options object

The options object passed to the function contains invalid options or values.

budget exceeded

The model consumption budget was exceeded.

unknown model "<model name>"

The specified model does not exist.

invalid language "<language>"

The specified language is not supported by the TRANSLATE function.

max tokens of <count> exceeded

The request exceeded the maximum number of tokens supported by the model (see Model restrictions).

all requests were throttled by remote service

The request has been throttled due to a high level of usage. Try again later.

Using Snowflake Cortex LLM functions with Python

Snowflake Cortex LLM functions are available in Snowpark ML version 1.1.2 and later. See Installing Snowpark ML for instructions on setting up Snowpark ML.

If you run your Python script outside of Snowflake, you must create a Snowpark session to use these functions. See Connecting to Snowflake for instructions.

The following Python example illustrates calling Snowflake Cortex LLM functions on single values:

from snowflake.cortex import Complete, ExtractAnswer, Sentiment, Summarize, Translate

text = """
    The Snowflake company was co-founded by Thierry Cruanes, Marcin Zukowski,
    and Benoit Dageville in 2012 and is headquartered in Bozeman, Montana.
"""

print(Complete("llama2-70b-chat", "how do snowflakes get their unique patterns?"))
print(ExtractAnswer(text, "When was snowflake founded?"))
print(Sentiment("I really enjoyed this restaurant. Fantastic service!"))
print(Summarize(text))
print(Translate(text, "en", "fr"))
Copy

You can also call an LLM function on a table column, as shown below. This example requires a session object (stored in session) and a table articles containing a text column abstract_text, and creates a new column abstract_summary containing a summary of the abstract.

from snowflake.cortex import Summarize
from snowflake.snowpark.functions import col

article_df = session.table("articles")
article_df = article_df.withColumn(
    "abstract_summary",
    Summarize(col("abstract_text"))
)
article_df.collect()
Copy

Note

The advanced chat-style (multi-message) form of COMPLETE is not currently supported in Python.

Using Snowflake Cortex LLM functions with Snowflake CLI

Snowflake Cortex LLM functions are available in Snowflake CLI version 2.4.0 and later. See Introducing Snowflake CLI for more information about using Snowflake CLI.

The following examples illustrate using the snow cortex commands on single values. The -c parameter specifies which connection to use.

Note

The advanced chat-style (multi-message) form of COMPLETE is not currently supported in Snowflake CLI.

snow cortex complete "Is 5 more than 4? Please answer using one word without a period." -c "snowhouse"
Copy
snow cortex extract-answer "what is snowflake?" "snowflake is a company" -c "snowhouse"
Copy
snow cortex sentiment "Mary had a little Lamb" -c "snowhouse"
Copy
snow cortex summarize "John has a car. John's car is blue. John's car is old and John is thinking about buying a new car. There are a lot of cars to choose from and John cannot sleep because it's an important decision for John."
Copy
snow cortex translate herb --to pl
Copy

You can also use files that contain the text you want to use for the commands. For this example, assume that the file about_cortex.txt contains the following content:

Snowflake Cortex gives you instant access to industry-leading large language models (LLMs) trained by researchers at companies like Mistral, Reka, Meta, and Google, including Snowflake Arctic, an open enterprise-grade model developed by Snowflake.

Since these LLMs are fully hosted and managed by Snowflake, using them requires no setup. Your data stays within Snowflake, giving you the performance, scalability, and governance you expect.

Snowflake Cortex features are provided as SQL functions and are also available in Python. The available functions are summarized below.

COMPLETE: Given a prompt, returns a response that completes the prompt. This function accepts either a single prompt or a conversation with multiple prompts and responses.
EMBED_TEXT_768: Given a piece of text, returns a vector embedding that represents that text.
EXTRACT_ANSWER: Given a question and unstructured data, returns the answer to the question if it can be found in the data.
SENTIMENT: Returns a sentiment score, from -1 to 1, representing the detected positive or negative sentiment of the given text.
SUMMARIZE: Returns a summary of the given text.
TRANSLATE: Translates given text from any supported language to any other.

You can then execute the snow cortex summarize command by passing in the filename using the --file parameter, as shown:

snow cortex summarize --file about_cortex.txt
Copy
Snowflake Cortex offers instant access to industry-leading language models, including Snowflake Arctic, with SQL functions for completing prompts (COMPLETE), text embedding (EMBED\_TEXT\_768), extracting answers (EXTRACT\_ANSWER), sentiment analysis (SENTIMENT), summarizing text (SUMMARIZE), and translating text (TRANSLATE).

For more information about these commands, see snow cortex commands.