Celeris-1 Decision
Diffusion-based decision model: yes/no, choice and score answers with probabilities and optional one-line reasons
Celeris-1 Decision is a decision model from Celeris, built on a diffusion language model. It evaluates noul (yes/no), choice and score questions about a state of text, JSON, images or a mix. It returns a probability for each outcome. An optional flag adds a one-sentence explanation per answer.
It is served at POST /v1/systemone and accepts the System One schema: existing Jev clients switch by changing the base URL and key. Celeris is operated by Marqo Inc.
Priced at $0.04 per 1M input tokens, with output free, explanations included.
Pricing: Per token usage
Celeris-1 Decision prices by model
Per 1M tokens, read off Celeris-1 Decision's own pricing page on the date shown.
| Model | Input / 1M | Cached / 1M | Output / 1M | Checked | Notes |
|---|---|---|---|---|---|
| Celeris-1 Decision | $0.04 | – | $0.00 | 15 Oct 2026 | Output free, explanations included. Images count as input tokens. |
Celeris-1 Decision Alternatives
Explore 147 products in the Inference APIs category. View all Celeris-1 Decision alternatives.
Kev
Open decision models from 0.8B to 27B, built on Qwen and compatible with TypeSafe's Jev API.
Laya
Open-source multilingual decision model: typed choices, scores and yes/no answers in one forward pass.
Liquid AI d1
Open-weight 3B decision model on LFM2.5-VL: typed answers over text, JSON and images in one forward pass
Nace.AI Drex
Jev-compatible decision models with open weights: a probability for every option from one pass, 128K context
ScaleDown Classify-1
Jev-compatible decision API for classification: label probabilities over text, images and documents up to 1M tokens
Work on Celeris-1 Decision? Feature it at the top of Inference APIs.
Is your product missing?