What features and capabilities does Manifest offer?
Manifest offers a single endpoint for all models, model fallbacks, full body logs, request Autofix, self-hosting, local model routing, and bring-your-own-key support. It is free, open-source, and allows easy parameter setup per route.
How is Manifest priced?
Manifest is a free, open-source product. It can be self-hosted at no cost, or used with its free tier. The pricing model includes subscriptions, pay-per-token, and support for local models and custom providers.
What is Manifest and what problem does it solve?
Manifest is an open-source LLM router and gateway designed to make AI agent calls reliable. It solves the problem of API failures in production, stating that 5% of LLM calls fail, most of which are avoidable. Its Autofix feature repairs bad requests on the fly to ensure zero downtime.
What is Manifest used for and in what situations?
Manifest is used to create reliable agents and workflows by providing a single endpoint to access all LLM models and providers. It is used when developers need to manage inference consumption, route requests across multiple providers, and handle provider outages or broken requests automatically.
Who is Manifest for?
Manifest is for AI builders, teams & startups, and platforms & providers. AI builders use it to connect plans and providers, teams manage consumption and tracking, and platforms help their users connect smoothly to models.
What is Manifest?
Manifest is an open-source LLM router and gateway for AI agents. It provides a single endpoint to access various LLM providers and models, with features for request fixing, fallbacks, and consumption management.