Supported AI models

View as Markdown

The models that Sigma uses for AI-powered features depend on the AI provider you set up when you configure AI features for your organization. You cannot choose which model Sigma uses, and Sigma might update the supported models over time.

The use of AI features is subject to the following disclaimer.

Reasoning models

Sigma uses a reasoning model for AI-powered features. The following table lists the reasoning model that Sigma currently uses for each provider. When Sigma plans to deprecate or replace a model, the relevant model and date are also listed:

AI providerModel statusModelDate
BigQuery
(warehouse-hosted)
CurrentGemini 3.6 Flash
DeprecatedGemini 2.5 ProAugust 10, 2026
PlannedGemini 3.1 Pro
Databricks
(warehouse-hosted)
CurrentClaude Sonnet 4.5
Deprecated
PlannedClaude Sonnet 5
Snowflake
(warehouse-hosted)
CurrentClaude Sonnet 4.5 (AWS-hosted accounts)
GPT-5.1 (Azure-hosted accounts)
DeprecatedClaude Sonnet 4 (AWS-hosted accounts)April 3, 2026
PlannedClaude Sonnet 5 (AWS-hosted accounts)
OpenAI
(external)
CurrentGPT-5.4
GPT-5.1
DeprecatedGPT-4.xFebruary 2, 2026
PlannedGPT-5.6
Gemini
(external)
CurrentGemini 3.6 Flash
DeprecatedGemini 2.5 ProAugust 10, 2026
PlannedGemini 3.1 Pro
Anthropic
(external)
CurrentClaude Sonnet 4.5
Deprecated
Planned
Amazon Bedrock
(external)
CurrentClaude Sonnet 4.5
Deprecated
PlannedClaude Sonnet 5
Claude Haiku 4.5

If you use Azure OpenAI, you can choose which model to use. Sigma recommends using the same model listed as Current for OpenAI.

Embeddings models

Sigma uses an embeddings model to create embeddings of your documents for semantic search. The following table lists the embeddings model that Sigma currently uses for each provider:

AI providerCurrently supported embeddings model
BigQuery (warehouse-hosted)gemini-embedding-001
Databricks (warehouse-hosted)GTE v1.5
Snowflake (warehouse-hosted)snowflake-arctic-embed-l-v2.0
OpenAI (external)text-embedding-3-small
Azure OpenAI (external)text-embedding-3-small
Gemini (external)gemini-embedding-001
Amazon Bedrock (external)Amazon Titan Text Embeddings

By default, Sigma uses your AI provider to generate embeddings. To use a different provider for embeddings, configure a separate Embeddings provider in Administration > AI settings and select another provider from this table. If you use Anthropic as your AI provider, you must configure a separate embeddings provider because Anthropic does not provide an embeddings model that Sigma can use. You cannot bring your own embeddings model.

When you update the embeddings provider, Sigma starts a reindex of searchable sources in the background.