Decision Models' s1-vision answers image questions with probabilities in 133ms
airesearch12 · x · 2026-11-16
Decision Models showed off the vision capability of its decision model s1-vision: send one photo plus one question, get back a typed answer with a probability. In the demo, "Is the parcel visibly damaged?" returned yes at 92% probability with a 133ms server-side response.
The company's site lists several Jev-compatible text/image/routing decision models, including REAX 1.3 (top 3 on the HF Decision Index, $0.400/1M input tokens), s1-pro ($0.103/1M), and s1-fast (97 decisions/sec on one GPU at 1/3 of s1-pro's price), mostly EU-hosted open-weight models.
More from Models
- Leaked 'Fable 5.5' / 'Model 4' is likely a new Claude pricing tier, not a new model — lyraxana · 2026-11-16
- It's only Monday and my weekly AI usage quota is nearly gone — gefei55 · 2026-11-16
- Alibaba says Qwen4-Max training data filtered to weaken offensive cyber and CBRN risks — terryyuezhuo · 2026-11-16
- Qwen4 Max jumps 9 points on Intelligence Index while cutting task costs 67% — 141_1337 · 2026-11-16
- Hy4 flash goes free with near-unlimited usage for two weeks: 1M context, multimodal, ZDR — udmrzn · 2026-11-16
- Claude quota hacks thread: match models to tasks, batch prompts, exploit Project caching — aitrendz_xyz · 2026-11-16