xiry.

AI · Web app

Cheaper Inference

OpenAI-compatible API marketplace routing model requests at discounted rates

From the Xiry library · Updated 10 October 2026

Visit cheaperinference.com ↗

Cheaper Inference is an AI routing marketplace that sells access to models from providers like OpenAI, Anthropic, and DeepSeek at discounts up to 76%. Developers switch upstream calls simply by pointing their existing OpenAI SDK base URL and API key to the platform. It features usage-based pricing with no commitments and offers optional Zero Data Retention (ZDR) routing.

What it's for

  • Lower token inference bills across production AI apps by changing two configuration values
  • Route prompts to top LLMs and media models through a unified, OpenAI-compatible API endpoint
  • Enforce strict privacy requirements using routes with Zero Data Retention (ZDR)
  • Monetize surplus AI model computing capacity by listing it on the marketplace

Features

  • Drop-in OpenAI SDK compatibility requiring only a base URL and API key update
  • Live discount catalog showing market rates, strikethroughs, and 24-hour token volume
  • Zero Data Retention (ZDR) routing options for sensitive workflows
  • Support for models across OpenAI, Anthropic, Google, Meta, DeepSeek, and more
  • Usage-based billing with deposits starting at $5 and no monthly commitments
  • Optional passthrough support for prompt caching where supported upstream
  • Marketplace feature allowing capacity holders to sell unused inference credits

Pricing

Paid. Usage-based pay-as-you-go; minimum deposit $5, no subscription required.

Details

Type
Web app
Category
AI
Topic
AI Model Routing
Open source
No
Platforms
API, Web
For
Developers and AI engineering teams looking to reduce LLM API inference costs.
Alternative to
OpenRouter, LiteLLM, Portkey
Tags
api, llm, inference, cost optimization, openai compatible, router

Why it was saved

  • @fabrice_mayrand, 31 August 2026: Cuts inference costs substantially without changing request format; useful for any product built on OpenAI or other providers.

Compare

Related

Updated