Avian.io

Fast AI inference billed per token. DeepSeek V3.2, Kimi K2.5, GLM-5, MiniMax M2.5 via OpenAI-compatible API. From $0.23/M tokens

Avian.io provides fast AI inference for developers who want to call open models through an OpenAI-compatible API and pay only per token. Models listed include DeepSeek V3.2, Kimi K2.5, GLM-5.1, and MiniMax M2.5, with DeepSeek V4 Pro announced as newly available, all reachable through a single API key. Switching from OpenAI means changing one line of code, the base URL.

The site says every model runs on NVIDIA B200 GPUs with speculative decoding, and it advertises production-grade speed with no rate limits. Coding is a focus: Avian works with Claude Code, Cursor, Cline, Kilo Code, and more than twenty tools in total. On the security side it cites SOC 2 approved infrastructure on Microsoft Azure, GDPR and CCPA compliance, and a no-data-stored policy, and it lists vision, search, and tool support among its capabilities. Although this directory files it under Marketing with startup-tools tags, it is a developer infrastructure product.

Pricing is pay-as-you-go: you add credits and are billed per token with no subscription, so it is a paid service, with a free start offered for getting going. Sign up, add credits, and point your existing OpenAI client at the Avian base URL.

Category: Marketing. Pricing: paid. Visit website

Alternatives to Avian.io in Marketing