AI/ML API for AI agents
Agents read AI/ML API at once. Changes and sends are held for a person's approval by default.
About AI/ML API
AI/ML API provides a suite of AI models and solutions for various tasks, including text generation, image processing, and more.
What an agent can do in AI/ML API
Every AI/ML API action is classed by the platform as a read, a change or a send. Reads run at once. Changes and sends are held for a person by default.
Runs at once (reads)
Looks things up in AI/ML API without asking first. 17 actions.
- List Thread MessagesTool to retrieve a list of messages from a specific thread. Use when you need to fetch conversation history or message content from an AI assistant thread.
- List ModelsTool to list all available AI models from the AI/ML API. Use when you need to retrieve the complete catalog of 400+ models including chat, image, video, voice, and other model types.
- List Models With DetailsTool to list all available AI/ML models with detailed information including pricing, features, and capabilities. Use when you need to discover available models or get comprehensive model metadata.
- Text Chat CompletionTool to generate text completions or chat responses using a specified LLM model. Use after assembling the conversation history to produce the next response.
- Get AssistantTool to retrieve details of a specific assistant by ID. Use when you need to fetch configuration, model settings, instructions, or available tools for an existing assistant.
- Get Billing BalanceTool to retrieve the current billing balance for the account. Use when you need to check available credits, balance status, or auto-debit configuration.
- Get Luma GenerationTool to fetch Luma AI video generation results by generation IDs. Use after creating a generation to check its status and retrieve the generated video URL when completed.
- Get MessageTool to retrieve information about a specific message by its ID. Use when you need to fetch details of a particular message in a thread.
- Get Response by IDTool to retrieve a previously generated model response by its unique ID. Use when you need to access details of a specific response, including its output, status, and usage statistics.
- Get RunTool to retrieve a specific run by ID from a thread. Use when you need to check the status, results, or details of a previously created run.
- Get Run StepTool to retrieve a specific run step by its ID within a thread and run. Use when you need detailed information about a particular step's execution status and results.
- Get ThreadTool to retrieve information about a specific thread by ID. Use when you need to fetch thread details, metadata, or available tool resources for an existing conversation thread.
Show all 17
- List AssistantsTool to list all assistants associated with the account. Use when you need to retrieve available assistants with pagination support.
- List BatchesTool to get the status or results of a batch processing job. Use when you need to check the progress or retrieve results of a previously submitted batch.
- List Luma AI GenerationsTool to fetch user's Luma AI video generations. Use when you need to retrieve a list of all Luma AI video generation tasks for the authenticated user.
- List RunsTool to list all runs for a specific thread. Use when you need to retrieve runs with pagination support.
- List Run StepsTool to list the steps in a run. Use when you need to retrieve and examine the execution steps of a specific run within a thread.
Held for approval by default (changes and sends)
Held for a person's approval by default. The approval is recorded and the action is sent once, never twice, with 10 seconds to undo. 13 actions.
- Create AssistantTool to create an AI assistant with configurable model, instructions, and tools. Use when you need to set up a new assistant that can maintain conversation context and use tools like code_interpreter
- Create MessageTool to create a new message in a thread. Use when you need to add a user or assistant message to an existing conversation thread.
- Create RunTool to create a run that executes an assistant on a thread. The assistant processes messages in the thread and generates responses based on its instructions and available tools.
- Create ThreadTool to create a new thread for conversation with an assistant. Threads store messages and maintain context across interactions. Use when starting a new conversation or when you need a fresh context f
- Cancel RunTool to cancel a run that is currently in progress. Use when you need to stop an assistant run before it completes.
- Delete AssistantTool to delete an assistant by ID. Use when you need to remove an assistant that is no longer needed.
- Delete MessageTool to delete a specific message from a thread. Use when you need to remove an unwanted or erroneous message from a conversation thread.
- Delete ThreadTool to delete a thread by its ID. Use when you need to remove an existing thread from the system.
- Submit Tool OutputsTool to submit tool outputs for a run that requires action. Use when a run has status 'requires_action' and needs tool call results to continue execution.
- Update AssistantTool to modify an existing assistant's properties including name, instructions, model, and tools. Use when you need to update an assistant's configuration or behavior after it has been created.
- Update MessageTool to modify metadata for a specific message in a thread. Use when you need to update message metadata such as tags or custom fields.
- Update RunTool to update a run's metadata with key-value pairs. Use when you need to attach or modify additional information for a specific run.
Show all 13
- Update ThreadTool to update thread metadata and tool resources in the AI/ML API. Use when you need to modify existing thread properties or attach resources.
Changes wait in Needs you
Changes and sends in AI/ML API are held for a person's approval by default. The approval is recorded, and the action is sent once, never twice, with 10 seconds to undo.
Related connectors
AI/ML API questions
Can an agent change data in AI/ML API on its own?
Not by default. Changes and sends in AI/ML API are held for a person's approval. The approval is recorded, and the action is sent once, never twice, with 10 seconds to undo.
What can an agent read in AI/ML API?
This connector lists 17 read actions. Reads run at once, without waiting for approval.
How does AI/ML API connect?
It connects with API key. Connectors are built and maintained by third-party developers; availability is confirmed during scoping.