Orchestrate teams of AI agents.One API call at a time.
LoopLlama is the backend for agentic workflows. Describe the project, we assemble the agent crew, run the loop, and return results — built for engineering teams shipping real work.
Usage-based pricing · Pay only for the steps your agents run
Everything you need to run agent teams in production.
LoopLlama is opinionated about the hard parts so your team can focus on the work.
Composable agent crews
Define roles, tools, and budgets. We handle planning, dispatch, retries, and merging the work.
Stateful workflows
Long-running jobs with checkpoints, resumable from any step. No glue, no babysitting.
Bring your own tools
Expose your APIs as tools. Agents call them with structured arguments and typed results.
Built-in observability
Per-step traces, token accounting, and replay. Debug a multi-agent run like a stack trace.
Model-agnostic
Route per-step to Claude, GPT, open models, or your own. Cost and latency aware.
SOC-ready foundation
Tenant isolation, audit logs, and PII redaction baked in. SOC 2 in progress.
From brief to delivered artifact in one API.
No agent framework to learn. No prompt graphs to maintain. Just describe the work.
Describe the goal
Send a project brief — natural language, structured fields, or both. We parse it into a typed workflow spec.
We assemble the crew
LoopLlama selects roles, tools, and a coordination strategy. You can override any of it.
Run, monitor, deliver
Stream progress events, fetch the final artifact, or poll a job ID. Re-run from any checkpoint.
One call. A team of agents.
The LoopLlama API exposes workflows as a single endpoint. Stream events as agents work, or wait for the final artifact.
- Typed SDKs for TypeScript, Python, and Go
- Idempotent submissions with job IDs
- Webhooks for completion and step events
- Bring tools via OpenAPI or simple JSON schema
import { LoopLlama } from "@loopllama/sdk";
const ll = new LoopLlama({ apiKey: process.env.LL_API_KEY });
const job = await ll.workflows.run({
goal: "Draft the Q3 product brief from these notes",
inputs: { notes_url: "https://..." },
crew: ["planner", "researcher", "writer", "reviewer"],
budget: { max_steps: 25, max_tokens: 200_000 },
});
for await (const ev of ll.events(job.id)) {
console.log(ev.step, ev.agent, ev.status);
}
const result = await ll.workflows.result(job.id);
console.log(result.artifact); // → markdown briefBuilt for the work your team is already doing.
Research & briefs
Synthesize sources into structured deliverables — competitive scans, market briefs, regulatory summaries.
Code modernization
Migrate, refactor, and document codebases with reviewer-in-the-loop crews that respect tests.
Customer ops
Handle multi-step support cases: classify, gather context, draft replies, escalate when needed.
Data pipelines
Investigate anomalies, propose fixes, and open PRs against your data infra repositories.
Questions about running agent teams.
- What is LoopLlama?
- LoopLlama is an orchestration API for teams of AI agents. You define a workflow with a crew of agents, then trigger runs over a simple REST API — LoopLlama handles planning, sequencing, model calls, persistence, and usage metering.
- How does LoopLlama orchestrate multiple AI agents?
- Each workflow defines a crew of agents with distinct roles. On a run, LoopLlama executes the crew step by step — each agent sees the original input plus the prior agents' output — and records every step with its tokens and timing for replay and debugging.
- Which AI models does LoopLlama support?
- LoopLlama is model-agnostic. You can route each step to Claude, GPT, open models, or your own, optimizing for cost and latency without rewriting the workflow.
- Can agents use my own tools and data?
- Yes. Expose your APIs as tools via OpenAPI or a JSON schema, and use built-in connectors like Gmail and Salesforce. Agents call them with structured arguments, and OAuth tokens are encrypted at rest.
- How does LoopLlama pricing work?
- Pricing is usage-based — you pay only for the workflow steps your agents run. A free Developer tier includes 10,000 steps per month, and paid plans add more steps, seats, and longer trace retention.
- Can a workflow pause for human approval?
- Yes. Human-in-the-loop gates let an agent pause for input or approval on sensitive actions and resume exactly where it left off, so a person stays in control without stalling the run.
Stop wiring agents. Start shipping.
Get an API key, run your first workflow in under five minutes, and pay only for the steps your agents run.