AI/ML API
for LuumenAI

Inspect assistants, threads, and runs — and change them as approved steps

Connect AI/ML API and Luumen can read your assistants, threads, messages, runs, run steps, batch jobs, and billing balance on request. When something needs to change — a new assistant, an updated instruction set, a stuck run canceled — Luumen shows the exact call first and waits for your approval. Authentication is by API key.

Book a DemoDownload Luumen

The AI/ML API toolbox

30 tools: 17 read, 13 write. Reads answer instantly. Writes require approval by default. Everything is logged.

  • ReadGet AssistantRetrieve details of a specific assistant by ID.
  • ReadGet Billing BalanceRetrieve the current billing balance for the account.
  • ReadGet Luma GenerationFetch Luma AI video generation results by generation IDs.
  • ReadGet MessageRetrieve information about a specific message by its ID.
  • ReadGet Response by IDRetrieve a previously generated model response by its unique ID.
  • ReadGet RunRetrieve a specific run by ID from a thread.
  • ReadGet Run StepRetrieve a specific run step by its ID within a thread and run.
  • ReadGet ThreadRetrieve information about a specific thread by ID.
  • ReadList AssistantsList all assistants associated with the account.
  • ReadList BatchesGet the status or results of a batch processing job.
  • ReadList Luma AI GenerationsFetch user's Luma AI video generations.
  • ReadList Thread MessagesRetrieve a list of messages from a specific thread.
  • ReadList ModelsList all available AI models from the AI/ML API.
  • ReadList Models With DetailsList all available AI/ML models with detailed information including pricing, features, and capabilities.
  • ReadList RunsList all runs for a specific thread.
  • ReadList Run StepsList the steps in a run.
  • ReadText Chat CompletionGenerate text completions or chat responses using a specified LLM model.
  • WriteCancel RunCancel a run that is currently in progress. Approval by default
  • WriteCreate AssistantCreate an AI assistant with configurable model, instructions, and tools. Approval by default
  • WriteCreate MessageCreate a new message in a thread. Approval by default
  • WriteCreate RunCreate a run that executes an assistant on a thread. Approval by default
  • WriteCreate ThreadCreate a new thread for conversation with an assistant. Approval by default
  • WriteDelete AssistantDelete an assistant by ID. Approval by default
  • WriteDelete MessageDelete a specific message from a thread. Approval by default
  • WriteDelete ThreadDelete a thread by its ID. Approval by default
  • WriteSubmit Tool OutputsSubmit tool outputs for a run that requires action. Approval by default
  • WriteUpdate AssistantModify an existing assistant's properties including name, instructions, model, and tools. Approval by default
  • WriteUpdate MessageModify metadata for a specific message in a thread. Approval by default
  • WriteUpdate RunUpdate a run's metadata with key-value pairs. Approval by default
  • WriteUpdate ThreadUpdate thread metadata and tool resources in the AI/ML API. Approval by default

One prompt, start to finish

What a governed AI/ML API run looks like inside Luumen.

Questions

How does LuumenAI connect to AI/ML API?

Authorize once with API token. Luumen lists the scopes each action needs before you approve the connection, and credentials never appear in the chat.

Can LuumenAI change things in AI/ML API on its own?

Read actions answer immediately. Anything that writes — cancel run, create assistant, create message, create run, and more — is shown as a plan and requires approval by default, including the 4 actions classified as destructive. Administrators configure that per tool, so you decide exactly which actions can ever run unattended.

Who gets access to the integration?

You decide. Actions are granted per agent, skill, and team, and per environment — production is not staging. Read access can be broad while writes stay narrow.

Is there an audit trail?

Every call to AI/ML API — read or write, approved or declined — is recorded with the actor, the input, and the result, and can be linked to the ticket or change record.

All integrations
Adobe Marketing Agent MCPAdobe Experience Platform context, sandbox switches, and async task statusAffindaRead parsed documents, collections, and annotations — change them with approvalAgent MailGive the agent its own mailbox for reading threads and sending updatesAirOps MCPRead AirOps brand context and AEO data, then change prompts as approved steps

Put AI/ML API to work with Luumen

Connect in minutes. Every action scoped, approved, and audited from day one.

Book a DemoDownload Luumen