.
Take back the control of your LLM calls
Manifest is an open source LLM gateway available on the cloud or on your own infrastructure.
- Budgets and rate limits
- Spend per engineer
- Bring your own providers
- Local models
Trusted by engineers who ship at
Manifest is an AI model gateway that you can trust
Model fallbacks
Automatically retry on another provider when one goes down.
Full body logs
Inspect the complete request and response body for every call.
Autofix
Broken requests get repaired on the fly, before your agent sees the error.
Local models
Route to Ollama, LM Studio, llama.cpp or any local server. Decide what goes to the cloud and what stays on your machine.
Bring your own key
Bring your own keys for every provider and pay usage directly. Manifest allows API keys from all popular providers and OAuth tokens for subscriptions.
Easy parameter setup
Set the right model and parameters per route without touching the code. Set AI model parameters visually.
Cost visualization
See every dollar spent, broken down by agent, key and provider.
SSO / SAML
Bring your own identity provider and manage team access through SSO and SAML.
Custom setup
Need something specific? We tailor the deployment and configuration to your team.
Don't get locked into a single provider
Simultaneously use different kinds of model or inference providers based on the query. Manifest does not limit you to a restricted list of providers.
API key providers
Bring your own key from all the main providers to get instant access to all their models. Pay by the usage directly to the provider.
Subscription providers
Already paying for a monthly subscription? Connect it to Manifest to use those quotas first, fallback to pay-as-you-go only when limits are exceeded.
Custom providers
Plug in any OpenAI-compatible or Anthropic-compatible provider that exists out there! We don't limit you to the providers that we know only.
Local models
Run open-weight models at home! Manifest handles Ollama, LM Studio and llama.cpp as first-class providers so you can run them on your own infrastructure.
One endpoint for the whole team. Start free.
Frequently asked questions
How do shared budgets work?
Set a cap per key, daily, weekly or monthly. When a key reaches it, Manifest stops routing on that key and sends an alert. Nobody discovers a $3,000 weekend on next month's invoice.
Can I see who spent what?
Give every engineer their own key. The dashboard breaks cost down by agent, key and provider, so you can see which model is eating the budget and who is calling it.
Do we need Enterprise to run a team?
Free and Pro are built for one person. Seats, SSO/SAML, audit logs and custom retention come with Enterprise, on our cloud or on your own servers.
Do you offer an SLA?
Enterprise includes a contractual uptime SLA, with on-premise, VPC or private cloud install. Free and Pro run on best effort.
How do we onboard the team?
Change one base URL. Manifest speaks both the OpenAI and Anthropic APIs, so Claude Code, OpenClaw, the OpenAI SDK and the Vercel AI SDK need no code change. Hand each engineer a key and they keep the setup they already have.
Can we keep our own provider keys?
Yes. Bring a key per provider and pay each provider directly. We take no cut of your tokens. Subscriptions work too (Claude, ChatGPT, GitHub Copilot and others), and the self-hosted version routes to Ollama, LM Studio or llama.cpp when a prompt should not leave your network.