AI task orchestration,
done right.
Submit async AI tasks and watch them execute live — with retries, model fallbacks, cost tracking, and human-in-the-loop checkpoints built in.
No spam. We'll email you when the beta opens.
Submit your first task in minutes
One call to submit — retries, model fallbacks, and cost caps are just fields in the schema. Stream progress back as it runs.
import { QueueAI } from "@queueai/sdk"
const queue = new QueueAI({ apiKey: process.env.QUEUEAI_API_KEY })
const task = await queue.submit({
name: "deep-research",
input: { question: "What changed in the AI dev tooling space in 2026?" },
model: {
primary: "anthropic/claude-opus-4-8",
fallback: "anthropic/claude-sonnet-4-6",
},
budget: { maxCostUsd: 1.50 },
retries: 3,
})
for await (const event of task.stream()) {
console.log(event.step, event.status, event.tokens)
}Built for how AI work actually runs
Live execution dashboard
Watch each step run in real time. Retries surface as they happen. Budget ticks up as tokens are consumed. Nothing is hidden.
AI-native primitives
Model fallback chains, token budgets, and cost caps are first-class fields — not afterthoughts bolted onto a generic queue.
Human checkpoints
Name the moments where you want a human to approve before execution continues. Built into the task schema, not hacked in later.
Learn the patterns first
Before you build with QueueAI, understand the orchestration patterns underneath it. Fan-out/fan-in, adversarial verification, cost-bounded research — all documented with code.
Browse the pattern library