Products

Infrastructure for test-time scaling, from the token service up to the agent harness.

  1. 01

    Chrono Infer

    OpenAI-compatible token service, tuned for both ends of the load range: fast tokens for one user and high volume for batch jobs.

    Available

  2. 02

    Chrono Engine

    Heterogeneous execution across GPUs and spatial chips, scheduled per workload.

    To be done

    Coming later
  3. 03

    HYA

    Agent harness that runs, checks and refines long-horizon tasks on top of fast inference.

    To be done

    Coming later