API

One endpoint.
Every agent, every model.

POST to /v1/chat/stream from any language that can make an HTTP request. Same payload, same response shape, whichever agent or provider is behind it.

Start free trial Explore a live demo

14 days free · No credit card · Bring your own model keys

The contract

Two calls to learn. That is the API.

A JSON body in, a reply out — or a stream if you want the tokens as they land. Nothing to install, nothing to generate before the first request.

Chat

One synchronous POST. The right shape for cron jobs, serverless functions and back-office scripts.

POST /v1/chat

{
  "agent": "support-agent",
  "message": "Where is order #10294?"
}

Stream

Server-Sent Events for chat interfaces. Tokens arrive as the model produces them, not after it finishes.

SSE /v1/chat/stream

data: "We"
data: " refund"
data: " within 30"

OpenAPI spec

Generate a typed client for your language straight from the spec. Nothing hand-written to keep in sync.

openapi.yaml

openapi: 3.1.0
paths:
  /v1/chat
  /v1/chat/stream
  /v1/agents

One key

Send it in the x-api-key header. No OAuth round trip, and no provider credentials scattered through your app.

Request header

x-api-key: sk-so-•••••••

No OAuth round trip. No SDK.

Behind the endpoint

The hard parts are already written. Just not by you.

Conversation history, retries, rate limits, tracing and model routing live on our side of the request. Your integration stays a single call.

Retries with backoff

A 503 from a provider is retried with exponential backoff before your code ever hears about it.

Retries

attempt 1+0.5s503

attempt 2+1.5s503

attempt 3200

Rate limits

Per-account quotas protect your bill and the downstream services your tools call.

Rate limit

Requests / min42 / 50

Returned on every response header.

Tracing on by default

Every request lands in an OpenTelemetry trace with latency, tokens and cost. No instrumentation to add.

req_8f2c

auth

retrieve

llm

send

Model routing

Move an agent from GPT-4o to Claude in the dashboard. The request your app sends does not change at all.

Active model

GPT-4o→Claude Sonnet

Zero lines of your code changed.

What teams say

“Every time you want to add AI to your app, you end up writing a ton of code — managing API keys, handling conversation history, engineering prompts, counting tokens. And then six months later a better model comes out and you have to refactor everything. SentientOne just removes all of that.”

Daniel Marsh · CTO

Pricing

Plans for every team size.
Fourteen-day trial, cancel anytime.

Starter

$19/month

For developers putting the first AI feature into a product.

Start free trial

1 agent · 10k requests/mo

1 MCP server per agent
Knowledge — 100 credits †
Embeddable chat widget
Analytics · 30 requests/min
14-day logs & chat history
Community support
Streamable HTTP endpoint
AI Workspace (team chat)

Pro

Enterprise

Custom

For organisations with compliance, scale or data-residency rules.

Talk to us

Unlimited agents & requests

Unlimited MCP servers & knowledge
Self-hosted / single-tenant option
Custom rate limits & log retention
SSO and organisational support
Security review assistance
Dedicated SLA & onboarding

Prices in USD. Every plan includes bring-your-own model keys, the agent chat interface, and a 14-day trial. † 1 credit = 1 FAQ · 5 credits = 1 document.

Get started

Add AI in one POST. Not one sprint.

Create an agent, copy the key, send a message. No SDK to install and no client to generate before the first reply streams back.

Create an agent
Copy your API key
POST a message
Read the stream

Start free trial See pricing

14 days free · No credit card required

One endpoint.Every agent, every model.