{"slug": "celeris", "title": "Celeris", "summary": "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.", "body_md": "Same tasks, same scoring. Celeris lands within a few points of the frontier on accuracy - and at more than 10x the speed.\n\nCurrent 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.\n\nThe result is an OpenAI-compatible API that delivers intelligent responses in milliseconds.\n\n``` python\nfrom celeris import Celeris\n\nclient = Celeris(api_key=\"sk-…\")\n\n# same call shape you already know\nstream = client.chat.completions.create(\n    model=\"celeris-1\",\n    messages=[{\"role\": \"user\",\n               \"content\": \"Explain a Kalman filter\"}],\n    stream=True,\n)\n```\n\nExperience the new paradigm of inference.", "url": "https://wpnews.pro/news/celeris", "canonical_source": "https://celeris.ai/", "published_at": "2026-07-25 21:00:13+00:00", "updated_at": "2026-07-25 21:22:31.004235+00:00", "lang": "en", "topics": ["artificial-intelligence", "large-language-models", "ai-products", "ai-infrastructure"], "entities": ["Celeris"], "alternates": {"html": "https://wpnews.pro/news/celeris", "markdown": "https://wpnews.pro/news/celeris.md", "text": "https://wpnews.pro/news/celeris.txt", "jsonld": "https://wpnews.pro/news/celeris.jsonld"}}