RunInfra

RunInfra, by RightNow and backed by Y Combinator, provides API access to open models designed for AI agents. Its service focuses on fast responses and lower input costs through cached context, with one API key compatible with the OpenAI and Anthropic SDKs. Developers can browse a model library that lists prices, capabilities, cache-hit rates, and output speeds, and explore models including DeepSeek V4.

1 Flash, GLM 5. 3 Flash, Nemotron 3. 5 Lightning 30B, Qwen3.

8 27B, and Ornith 1. 5 35B. Listed capabilities include tool calling, JSON mode, streaming, and image input, depending on the model.

RunInfra also offers session-aware routing, which directs repeat calls to the replica holding an agent’s conversation context, helping preserve warm caches without requiring a routing hint. Cache hits are reported in usage and billed at the cached rate, though the company describes them as best effort. The platform publishes cache measurement methodology and provides dashboards for account access and usage