Celeris

Celeris, a new inference architecture for language models built on diffusion, delivers an OpenAI-compatible API that achieves accuracy within a few points of frontier models at more than 10x the speed, generating intelligent responses in milliseconds by overcoming the sequential latency of autoregressive models.

Same tasks, same scoring. Celeris lands within a few points of the frontier on accuracy - and at more than 10x the speed. Current autoregressive models generate one token at a time. Every token depends on the one before it, making latency fundamentally sequential. Celeris is a new inference architecture for language models, built on diffusion, that achieves latency and quality previous diffusion systems could not. The result is an OpenAI-compatible API that delivers intelligent responses in milliseconds. python from celeris import Celeris client = Celeris api key="sk-…" same call shape you already know stream = client.chat.completions.create model="celeris-1", messages= {"role": "user", "content": "Explain a Kalman filter"} , stream=True, Experience the new paradigm of inference.