How can we help?
Ask in your own words. The answer comes from our guides and links to the articles it used. If you need a person, our team can join the chat.
Browse by topic
Guides
All 42 articles- What is ModelRiver, and why not just call OpenAI directly?ModelRiver is one API in front of 13 AI providers, with failover, structured outputs, logs and spend limits. See when it beats calling a provider directly.
- How is ModelRiver different from a regular AI gateway?An AI gateway routes calls to providers. ModelRiver also delivers results over WebSocket, runs backend pipelines, sends webhooks and remembers chats.
- How is ModelRiver different from an orchestration framework like LangChain?Orchestration frameworks are libraries you run yourself. ModelRiver is a hosted service joining a gateway, pipelines, webhooks, WebSocket delivery and sessions.
- What's the difference between a project and a workflow?A ModelRiver project holds your providers, keys and logs. A workflow is a named AI setup your app calls. Learn how they fit together and how to organize them.
- Can I try ModelRiver without paying for AI calls?Try ModelRiver for free: the Free plan needs no card, and test mode returns sample responses without calling a provider. Here's how to test your integration.
- Does saving a workflow change what production uses?No. Saving changes only the Draft; production runs the Live version until you publish. Learn how ModelRiver Draft, Live, publish and rollback work.
- What happens to my app when an AI provider goes down?Add backup models to a ModelRiver workflow and requests fail over automatically when a provider errors, times out or rate-limits. Here's how failover works.
- How do I make sure the AI always returns valid JSON?Attach a structured output (JSON schema) to a ModelRiver workflow and every response is checked against it, with backups tried if it fails. Here's how.