Celeris
Celeris 1
Celeris 1 is a diffusion language model built for ultra-low-latency classification, extraction, judging, query rewriting, and other short structured responses.
Model facts
- Context window
- 8K tokens
- Maximum output
- 8K tokens
- Cheapest paid input
- $2 per 1M tokens
- Cheapest paid output
- $6 per 1M tokens
- Open weights
- no
- Providers
- 1
Capabilities and modalities
text.
Artificial Analysis benchmarks
Compare this model on the full LLM benchmark leaderboard.
| Configuration | AAI | Coding | Math | Output tokens/s |
|---|---|---|---|---|
| default | 12.4 | 14.4 | not measured | 2046.9 |
Observed price history
- 2026-08-24: $2 input / $6 output per 1M tokens
API providers
- NanoGPT — model id celeris-1; $2 input / $6 output per 1M tokens; provider documentation