O

OpenRelay

Infrastructure service

by OpenRelay

Inference and GPU compute platform with an OpenAI- and Anthropic-compatible endpoint that routes each request to a site by latency, cost and availability, billed per token on demand, as dedicated capacity with an SLA, or in half-price batches, plus rentable GPUs such as the H100 SXM at $2.60 an hour and RTX 4090 at $0.35 an hour billed per second, driven from the orl CLI or MCP server; self-serve with a card and deposit, or by demo.

Find alternatives to OpenRelay

What it does

• GPU cloud
• Model inference
• Model hosting

Features 4

• One inference endpointAn OpenAI- and Anthropic-compatible endpoint billed per token with no minimum.
• Rentable GPUsRoot over SSH on a rented GPU virtual machine with a persistent volume, billed hourly.
• CLI and MCP serverThe orl CLI wraps the API and orl mcp serve exposes the platform as MCP tools.
• Batch at half priceSubmit a JSONL file that runs as capacity frees up at half the on-demand rate.

Sources

Pricing

Pricing page · 20 Oct 2026

Pricing page: GPU rates include H100 SXM 80GB $2.60 /hr, RTX 4090 24GB $0.35 /hr and RTX 3090 24GB $0.18 /hr, with per-second billing.

Description

Official documentation · 20 Oct 2026

Home page: "Buy inference per token, a month or a quarter at a time with an SLA, or in batches at half price."

What it does

Official documentation · 20 Oct 2026

Home page: "Buy inference per token, a month or a quarter at a time with an SLA, or in batches at half price."

Constraints

Official documentation · 20 Oct 2026

Derived from the security facts on this page, from the same source.

Security

Official documentation · 20 Oct 2026

Security page lists Owner, Admin, Member and Viewer organization roles; it names no certification and the pages read do not state a training policy for customer content.

Sold within

Pricing page · 20 Oct 2026

Self-serve: the home page says Add a card and make a deposit to start; it also offers Book a demo.

Name

Official documentation · 20 Oct 2026

Name as published on the product page.

URL

Official documentation · 20 Oct 2026

Product page, read 2026-10-20.

Deployment

Official documentation · 20 Oct 2026

The pages read state no deployment model beyond the vendor-run service; saas is recorded as the vendor-operated reading.

Interfaces

Official documentation · 20 Oct 2026

Home page: an API endpoint, the orl CLI, orl mcp serve as MCP tools, and a sign-in dashboard.

Open OpenRelay in the directory

Something wrong here? Send a correction — quote this product id: openrelay-inference-compute.