Can I call an LLM directly from the browser with Behest?

Yes. Most LLM providers block browser requests because they do not support CORS, which forces you to build a server proxy. Behest adds per-project CORS allowlists so your browser code can call the API directly. You set your allowed origins, and Behest applies auth, PII scrubbing, and rate limits in the request path.

Does adding AI this way include conversation memory and rate limiting?

Yes. Behest stores conversation history per user and per session in Redis, so multi-turn chat works without you managing context. It also enforces three tiers of rate limiting — per-IP, per-project, and per-user — so abusive or runaway traffic is capped before it reaches your LLM.

Is 'Add AI to Your App' the main Behest product?

No. Add AI to Your App is the developer capability of the Behest AI Token FinOps platform — the control center for your AI spend. This browser-callable backend is the entry point; teams scale into per-session cost attribution, budget enforcement, and AI Governance as their AI usage grows.

What does it cost to start adding AI to my app?

There's a free tier to build and evaluate, and an Enterprise license for production. Behest is a SaaS license, not a per-token markup, so you pay for the platform rather than a surcharge on every model call. You can bring your own provider keys or use a Behest-managed default.

Developer Edition

Add AI to Your App. No Backend Required.

A capability of the Behest AI Token FinOps platform.

Last updated: April 27, 2026

Behest is the AI backend you don't have to build. CORS handled, auth included, PII scrubbed, prompts defended — call from your frontend, ship in minutes.

Add AI to Your App is Behest's developer surface: an OpenAI-compatible /v1/chat/completions endpoint you call from the browser with CORS allowlists, JWT auth, Presidio PII redaction, Sentinel prompt-injection checks, Redis conversation memory, per-session cost capture, and token budgets — no bespoke Node proxy required.

Start the 5-minute Quickstart View on GitHub

✓ Includes employee token usage tracking

Everything between your app and your LLM

One API call. Behest handles the rest — so you focus on the features your users see.

CORS handled

Configure your origins. Call the API from the browser.

Auth included

Multi-tenant JWT + per-project API keys, RBAC built in.

Conversation Memory

Persistent context across sessions — Redis sorted sets, per user.

PII Shield

Microsoft Presidio scrubs sensitive data before the LLM sees it.

Sentinel

Prompt-injection defense in-line on every request.

AI Token FinOps

Per-session cost capture on every call; full attribution dashboards and cost-center rollups on Enterprise.

Build here. Scale with FinOps and Governance.

The same gateway developers use on this page powers enterprise programs: finance-grade attribution and caps on AI Token FinOps, plus policy and audit controls on AI Governance.

Cost & allocation

AI Token FinOps

Per-session spend visibility, cost-center rollups, and budgets that enforce before the invoice — for CFO and platform teams.

Risk & compliance

AI Governance

Model allowlists, EU AI Act–aligned contracting hooks, NIST AI RMF mapping, PII Shield, Sentinel, and audit trails.

Built-in AI Token FinOps

Enforce hard budgets, attribute costs per session, and get complete visibility into your enterprise AI spend.

Budget Limits

Cost Attribution

Learn more

Call from the browser. Ship today.

No backend proxy. Same OpenAI-compatible request shape your code already speaks.

your-app.tsx — Add AI to your existing app

// No backend proxy needed — CORS handled
// Call Behest directly from your browser

const response = await fetch(
  "https://your-project.behest.app/v1/chat/completions",
  {
    method: "POST",
    headers: {
      "Authorization": `Bearer ${userToken}`,
      "Content-Type": "application/json",
      "X-End-User-Id": userId,
      "X-Session-Id": sessionId,
    },
    body: JSON.stringify({
      model: "gemini-2.5-flash",
      messages: [{ role: "user", content: message }],
    }),
  }
);

// Conversation Memory persists across sessions
// PII scrubbed before reaching the LLM
// Per-user rate limits + token budgets enforced
// Per-session cost attribution captured automatically

The AI Backend You Don't Have to Build

One API Call. Everything Handled.

Stop building auth, rate limiting, CORS proxies, and memory management. Behest runs the complete AI backend — in our cloud or yours. You write the frontend.

Zero AI backend code to write or maintain
Call directly from your browser — CORS handled
PII scrubbed and prompts defended automatically
Self-hosting option — your data stays yours

CORS-Ready

Call from your browser — no backend proxy needed

Auth Built In

Multi-tenant JWT + API keys with tenant isolation

Memory Included

Persistent conversation context across sessions

PII Protected

Automatic detection and scrubbing before it reaches the LLM

Prompts Defended

Sentinel blocks injection attacks with pattern detection

Rate Limited

Three-tier limits per IP, project, and user

Fully Observable

Traces, metrics, logs — correlated in Grafana

Flexible Deployment

SaaS in our cloud, or deploy in yours — you own the keys and egress

Frequently asked questions

How do I add AI to my web app without building a backend?: Point your frontend at your Behest project URL. Behest handles CORS natively, so your React, Vue, Svelte, or vanilla JS app can call an OpenAI-compatible /v1/chat/completions endpoint directly — with JWT auth, per-project API keys, and rate limiting enforced on every request. No bespoke backend proxy required.
Can I call an LLM directly from the browser with Behest?: Yes. Most LLM providers block browser requests because they do not support CORS, which forces you to build a server proxy. Behest adds per-project CORS allowlists so your browser code can call the API directly. You set your allowed origins, and Behest applies auth, PII scrubbing, and rate limits in the request path.
Does adding AI this way include conversation memory and rate limiting?: Yes. Behest stores conversation history per user and per session in Redis, so multi-turn chat works without you managing context. It also enforces three tiers of rate limiting — per-IP, per-project, and per-user — so abusive or runaway traffic is capped before it reaches your LLM.
Is 'Add AI to Your App' the main Behest product?: No. Add AI to Your App is the developer capability of the Behest AI Token FinOps platform — the control center for your AI spend. This browser-callable backend is the entry point; teams scale into per-session cost attribution, budget enforcement, and AI Governance as their AI usage grows.
What does it cost to start adding AI to my app?: There's a free tier to build and evaluate, and an Enterprise license for production. Behest is a SaaS license, not a per-token markup, so you pay for the platform rather than a surcharge on every model call. You can bring your own provider keys or use a Behest-managed default.

Zero to first API call in under 5 minutes.

Free to build and evaluate. No SDK required — standard fetch from any frontend.

Start the 5-minute Quickstart Read the docs

✓ Includes employee token usage tracking