When reliability matters
Automations that draft support replies, summarize incidents, classify intake, or enrich records need a clear outcome. Silent failure creates manual cleanup work.
ReqRun gives each LLM request a durable lifecycle so your automation can continue, retry, or ask a human to inspect a failed request.
The product boundary
ReqRun does not provide a no-code workflow canvas, callback routing, or multi-provider execution in v1.
It is the reliable request layer you call from your own automation code.
const response = await reqrun.chat.completions.create({
model: "gpt-5-nano",
messages: [{ role: "user", content: "Run the agent task." }],
wait: false,
idempotency_key: "agent-task-991",
});
if (response.object === "chat.completion.async") {
const request = await reqrun.requests.get(response.id);
console.log(request.status, request.attempts);
}