MmkecoBrowser & computer use

Cerebellum-2B

Non-Autoregressive AI Agent Decision Model. Open-source SOTA alternative to TypeSafe Jev. O(1) Tool Routing & DOM Automation on Qwen3.5-2B .

Explore repositoryVisit website

How it uses Jev

🧠 Cerebellum-2B (小脑-2B) 25ms 极速非自回归 AI Agent System 1 决策引擎 ⚡ TypeSafe Jev 与 KEV 开源 SOTA 对标首选 • $O(1)$ 单次前向动作路由

Source: community catalog. Details and metrics reflect the published community snapshot and are not a performance endorsement. Documentation details extracted from GitHub · the README · the project website.

Keep exploring

Similar to Cerebellum-2B

Browse all
von project preview

von

Jev alternatives & open source
The open-source System One decision model. Sub-15ms, non-autoregressive, local drop-in alternative to TypeSafe Jev.
WMade bywfzyx610
Jev Ultrafast project preview
Image 1 of 2

Jev Ultrafast

Browser & computer use
A browser agent that turns a single goal into a sequence of lightning-fast actions.
BMade bybrowser-use19.5k
jev-browse (0x7067) project preview

jev-browse (0x7067)

Browser & computer use
Browser automation with Jev (TypeSafe) as decision model
0Made by0x70673
jev-browser (KesavanKing) project preview

jev-browser (KesavanKing)

Browser & computer use
Local browser automation UI that uses TypeSafe Jev to choose bounded page actions and a text model only for field values.
KMade byKesavanKing0
Verdict-open-jev project preview

Verdict-open-jev

Jev alternatives & open source
Non-autoregressive decision engine on ModernBERT (151M) with calibrated uncertainty (RLCD), TypeSafe AI Jev benchmark audit, and in-browser WebGPU playground
HMade byHeman10x-NGU94
open-spark-jev project preview

open-spark-jev

Benchmarks & research
Open-source, local decision models inspired by TypeSafe’s Jev and System One - built on Qwen3 for NVIDIA DGX Spark.
AMade byabhishek08515
open-jev-typed-decision-engine project preview

open-jev-typed-decision-engine

Jev alternatives & open source
Open reproduction of TypeSafe Jev: a 150M typed decision engine (noul/choice/score in one non-autoregressive pass, calibrated confidence). 0.697 vs Jev's 0.727, 2.5x better calibrated, 4x faster, free. Trains on a Colab T4 in 30 min.
IMade byintikhab4943
openpave-jev project preview

openpave-jev

Benchmarks & research
PAVE skill for non-autoregressive decision models (Jev / Laya / TypeSafe): typed Choice, Score and Noul answers with calibrated probabilities
CMade bycnrai2
jev-reward-model-evaluation project preview

jev-reward-model-evaluation

Benchmarks & research
Jev 1.13 reward-model evaluation across 8 benchmark tracks, with an interactive report and 54-row SOTA comparison
GMade bygoya41404