Retool AI providers and models
Learn about the different AI providers and models that Retool supports.
AI providers use models that are trained on specific types of data to make relevant decisions and perform tasks. For example, you would interact with one type of model to generate chat responses and a different model to generate images. AI models are also known as LLMs (large language models).
In general, a model is a collective group of model instances, each of which vary in terms of functionality or data. For example, there are different instances of OpenAI's GPT-4 model, such as gpt-4 and gpt-4-turbo. You can specify the exact model instance when configuring Retool AI.
Retool supports a number of AI providers and their models for use with AI actions and agents.
AI tokens
Use OpenAI's tokenizer tool to calculate how many tokens would result from a passage of text.
AI models use tokens, rather than characters or bytes, to represent the commonality of sequences and usage. In general, a single token can equate to approximately four characters.
The text you provide in an AI query is converted by the model into tokens, after which it can generate the necessary response.
Retool-managed and self-managed models
LLM interactions work by making API requests with an AI provider. As with most other APIs, these requests require authorization credentials, such as an API key. Retool provides access to certain AI models using Retool-managed connections that use Retool's own API keys.
A Retool-managed LLM connection enables you to start building and testing with AI models straight away without needing any additional tools or services. Once you're ready to use your AI actions or agents in production, you then need to provide your own authorization credentials. The connection then becomes self-managed.
For instance, you can immediately start building an agent using GPT-4o-mini with the Retool-managed OpenAI connection. Once your agent is ready for use, you then update the OpenAI connection and provide your own OpenAI API key.
Some AI models are only available for use as self-managed connections.
Available AI providers and models
Retool supports a number of AI providers and their models for use with AI actions and agents. Agents only support models that allow tool creation. Refer to the following information to learn which models are supported with AI actions or agents.
OpenAI
Retool-managed and self-managed OpenAI models are available for use in AI actions and agents.
| Model | AI actions | Agents |
|---|---|---|
| GPT-5 1 | ||
| GPT-5-mini | ||
| GPT-5-nano | ||
| GPT-4.1-mini | ||
| GPT-4.1 | ||
| GPT-4o-mini | ||
| GPT-4o | ||
| GPT-4 | ||
| GPT 3.5 Turbo | ||
| o3 | ||
| o3-mini | ||
| DALL·E | ||
| o1 | ||
| ChatGPT-4o-latest 2 |
Footnotes
Anthropic
Retool-managed and self-managed Anthropic models are available for use in AI actions and agents.
| Model | AI actions | Agents |
|---|---|---|
| Claude Sonnet 4.5 | ||
| Claude Sonnet 4 | ||
| Claude Opus 4 | ||
| Claude Sonnet 3.7 | ||
| Claude Haiku 3.5 | ||
| Claude Sonnet 3.5 | ||
| Claude 3 | ||
| Claude 2.1 | ||
| Claude 2 | ||
| Claude Instant 1.2 |
Google Gemini
Self-managed Google Gemini models are available for use in AI actions and agents.
| Model | AI actions | Agents |
|---|---|---|
| Gemini 2.5 Pro | ||
| Gemini 2.5 Flash | ||
| Gemini 2.0 Flash |
Azure OpenAI
Self-managed Azure OpenAI models are available for use in AI actions and agents.
Rather than selecting from a predefined set of models, you create and deploy an Azure OpenAI service with a specific model. You then provide the endpoint and model name when configuring the API credentials. Refer to Connect to Azure OpenAI for how to get started.
Retool charges a flat-rate hourly fee with the self-managed LLM key model price for Azure OpenAI.
Amazon Bedrock
Self-managed Amazon Bedrock models are available for use in AI actions only.
Amazon Bedrock is a managed service that can be configured for different AI models, such as Anthropic or Cohere models. The models available for use with Retool AI depend on the models you configure for Bedrock.
Baseten
Retool-managed and self-managed Baseten models are available for use in AI agents only.
| Model | AI actions | Agents |
|---|---|---|
| DeepSeek-V3 | ||
| DeepSeek-r1 | ||
| Llama 4 Maverick | ||
| Llama 4 Scout |
Custom AI provider
Self-managed custom AI provider models are available for use in AI actions and agents.
You can connect a custom AI provider to Retool that adheres to an OpenAI, Anthropic, Google, or Cohere schema.
AI usage rate limiting
Retool imposes the following rate limits on Retool-managed key usage. Rate limits are per organization, per hour (over a rolling time window).
| Feature | Hourly rate limit |
|---|---|
| Retool Agents | 20,000,000 tokens |
| Assist | 100,000,000 tokens |
| All other AI functionality 1 | 250,000 tokens |
This rate limit is intended to prevent abuse of resources and to limit Retool-managed key usage to non-production instances. Each AI provider also imposes their own rate limits, which also apply to Retool users.
If you encounter a rate limit imposed by Retool or by an AI provider, you will receive a 429 error with a message describing the kind of rate limit that was hit.
Rate limits are cumulative, and they encompass all AI usage. For example, if you are using Vectors, you may reach this limit very quickly. These rate limits apply to any customer using Retool-provided keys, regardless of deployment method or plan type. Retool is unable to make exceptions to or raise this rate limit.
If you do not wish to be subject to Retool-imposed rate limits, or are ready to use AI in production, configure AI platforms to use your own API credentials. Self-managed connections are not subject to Retool-imposed rate limits.
Model usage and pricing
Retool's pricing for AI model usage differs between Retool features.
Assist pricing
Usage for Assist is measured by credit. For a limited time, free plans can use up to 100 credits for free, and paid plans (Team, Business, and Enterprise) can use unlimited free credits.
Refer to the Assist billing and usage support documentation for more information.
Retool Agents pricing
Usage for Retool Agents is measured by hour, and the hourly rate differs for Retool-managed LLMs and self-managed LLMs. Refer to the Agents billing and usage support documentation for more information.
- Retool-managed LLMs
- Self-managed LLMs
Retool provides access to OpenAI, Anthropic, and Baseten AI models for Agents through a Retool-managed connection.
Retool uses an hourly rate for billable usage of managed LLM connections. The exact cost varies depending on provider costs, as shown in the table below.
You can provide your own authorization credentials for each AI provider and the models supported by Retool. Retool also supports the use of custom AI providers, allowing you to make use of many more models.
Once you are ready to use Retool Agents in a production environment, you must provide your own credentials. This ensures that your agents aren't restricted by the Retool-managed connection limits.
To simplify billing, Retool charges a flat-rate hourly fee when using a self-managed LLM connection, as shown in the Self-managed LLM/key entry of the table below.
Refer to Agent usage for more information about Billing and usage for Retool Agents.
Footnotes
-
Including, but not limited to: AI actions, Ask AI, AI-generated READMEs, Retool Vectors, and the LLM Chat component. ↩