Jeff v1.4: a 0.8B decision model with 34 open adapters, 90%+ accuracy, 2ms inference
Usual_Maximum7673 · reddit · 2026-11-06
The author releases Jeff v1.4, a local 0.8B model with LoRA adapters trained to make decisions and return calibrated probabilities.
What's new in v1.4:
- 34 open-weight decision adapters with an automatic router
- 90%+ accuracy on supported domains
- 2ms inference per decision (40ms via hosted API incl. network)
- 5 SDKs: Python, Node.js, Rust, C#, Java
- Runs on both CUDA and MLX
- Playground, docs and worked examples
The core pitch: turning decisions into a software primitive — like expressions, but for unstructured data. Code, weights and adapter catalogue are all open-sourced (GitHub/Hugging Face/jeffhub.ai).
More from Models
- Bittensor miners train open-source decision model REAX 1.3 — const_reborn · 2026-11-06
- Bittensor subnet 92 releases REAX 1.3, an open-source decision model blending miner models — const_reborn · 2026-11-06
- Kimi K3.1 Pro scores 9.9% on agentic WeirdML v3, ahead of Kimi-K3 but behind Claude Haiku 5.5 — scaling01 · 2026-11-06
- Rumor: GLM-5.5 to come in 32B and 122B sizes, community figure says no — TheZachMueller · 2026-11-06
- User screenshot suggests Anthropic's weekly Claude limits feel effectively unlimited — StewartalsopIII · 2026-11-06
- Leonie Shares Full Slide Deck on Building Decision Models, Still Early Days — helloiamleonie · 2026-11-06