Decido-1: open 26B MoE decision model beats Jev 1.13, runs CPU-only on laptops
gauravapiscean · reddit · 2026-11-12
A developer released Decido-1, open small models purpose-built for yes/no and multiple-choice decisions inside agents, returning per-option probabilities instead of generated text.
- Decido-1: 26B MoE (4B active), scores 85.4 across the six benchmarks Jev 1.13 is evaluated on (Jev: 83.9); 4s CPU / 1.1s Metal on an M4 Pro for a 1k-token doc with 5 questions
- Decido-1-Max (reasoning-first): 89.2, but slow on laptops; 1.7B and 0.6B variants score 73.0 / 62.2 at 0.4-0.6s CPU latency
- Q80 GGUFs on llama.cpp, bit-identical CPU outputs; option-order answer-flip rate 4.8% vs Jev's reported 14.6%
- Weights on Hugging Face, pip install decido-ai, plus a no-signup playground
More from coding & agent
- United Airlines to Ban Personal AI Agents From Booking Flights for Customers — Polymarket · 2026-11-12
- Agentic Search explained: a 4-search loop costs about 5x one search — and queries leak to more companies — adnan_hashmi · 2026-11-12
- AWS says thousands of autonomous AI agents wrote every line of AWS Emulator — SumitGup · 2026-11-12
- freeCodeCamp launches free 2-hour course on agentic coding with Codex — adnan_hashmi · 2026-11-12
- A criminally underrated resource comparing tool-calling reliability across models and providers — ricklamers · 2026-11-12
- Organizing 55k data points is easy: enterprise data prep for agents is the hard part — BenBajarin · 2026-11-12