OpenRouter unified LLM API

Offering overview

Instead of hosting GPUs, this New York company aggregates inference from more than eighty upstream providers behind a single API key and credit balance. Over 500 text, image, video and audio models are reachable with OpenAI-compatible requests, and the router can sort providers by price, latency or throughput, fall back to alternates when one degrades, and honor per-request provider ordering or exclusions. Privacy controls let customers refuse providers that log or train on prompts, and many models expose a free variant capped at 50 requests per day (1,000 after buying ten dollars of credits). Model families available include Llama, Mixtral and Mistral, GPT, Claude, Gemini, DeepSeek, Qwen and hundreds more, making it a common choice for app builders who want to swap models without rewriting integrations. Token costs are passed through from the underlying providers with no markup; revenue comes from a 5.5 percent fee (minimum $0.80) on credit purchases and a 5 percent surcharge on bring-your-own-key usage above a $25,000 monthly allowance. There are no subscriptions.

Technical specifications

Containers

Unknown

Kubernetes

Unknown

Root access

Unknown

API access

Unknown

Autoscaling

Unknown

Persistent processes

Unknown

AI capabilities

Model serving

Llama 3.1, Mixtral 8x7B, GPT-4o, Claude, Gemini, DeepSeek, Qwen, Mistral

AI frameworks

Not stated

Inference endpoints

Unknown

Fine-tuning

Unknown

Vector database

Unknown

Model catalog

Unknown

Monitoring

Not stated

Evaluation

Not stated

Security & compliance

Geography

Provider headquarters are separate from offering infrastructure and availability.

Available plans

Pricing can change. Promotional prices are shown separately from standard prices.

No active plans are recorded for this offering.

Questions & answers

What does OpenRouter provide?

OpenRouter aggregates inference from more than 80 providers through one API key and credit balance, offering access to over 500 text, image, video and audio models.

Who is OpenRouter useful for?

It serves app builders who want to swap models without rewriting integrations, using OpenAI-compatible requests.

How can developers control provider routing?

Developers can sort providers by price, latency or throughput, set per-request provider ordering or exclusions, and fall back to alternate providers when one degrades.

Can customers exclude providers that log prompts?

Privacy controls let customers refuse providers that log prompts or use them for training.