Celeris-1 claims near-GPT-5 intelligence with 157 ms latency and 1,280 tok/s
alejandroll10 · x · 2026-07-25
Celeris.ai says its new general-purpose model, Celeris-1, uses a diffusion-based inference architecture instead of standard autoregressive decoding.
- The company claims 157 ms p50 latency, about 15× faster than GPT-5-mini and 17× faster than GPT-5.
- On MMLU-Pro, it reports 75.9%–76% accuracy, close to GPT-5-mini and GPT-5.
- In a tokens-per-second comparison, it says Celeris-1 reaches 1,280 tok/s versus 144 tok/s for Gemini 3.5 Flash Lite.
- The model is available now.
Related event: Celeris-1 Launches with Diffusion-Based Low-Latency Inference(3 posts)→
More from Models
- Fern's "context is intelligence" thesis: prediction keeps improving out to 10.2M tokens — bneyshabur · 2026-10-26
- Mirendil launches Fern, a from-scratch pretrained model with up to 10M-token context — bneyshabur · 2026-10-26
- Users spot a dedicated translation UI appearing in Claude — BLCNYY · 2026-10-26
- GrokBot user: flight booking and shopping all blocked, agent can't do anything useful — lucasmeijer · 2026-10-26
- Vending-bench-style eval: Sol 5.6 appears the most aggressive contract-violator — xuanalogue · 2026-10-26
- Fable 5.5 launch rumors persist for third week, but no frontier model drops this month — patience_cave · 2026-10-26