Study of 2,170 GitHub projects maps how the fast, low-cost Jev decision model is used in the wild
CUHK-CSE · hf · 2026-09-28
CUHK researchers analyzed 2,170 public GitHub projects (as of Sep 22, 2026) built around Jev, a fast, low-cost decision model that answers natural-language questions with choices, binary judgments, and scores.
- Ecosystem growth: rapid early growth, both new projects and integration into existing repos.
- Usage patterns: attribute judgment and scoring are widely used; action selection, content filtering, and model/tool selection vary by domain. Jev acts as a reusable decision component whose role shifts with the surrounding workflow.
- Attention mismatch: public attention concentrates on routing and interface agents and does not track project counts.
The study offers a quantitative view informing the design and evaluation of general-purpose decision models.
More from coding & agent
- DIY harness × model benchmark: Claude Code hits 100% while deepseek-v4.1-flash matches 96% in half the time — dh7net · 2026-09-28
- ComfyVault dedupes model files across ComfyUI installs via symlinks — ruashots · 2026-09-28
- Agent failed for weeks on a problem StackOverflow answered instantly — itsOmSarraf_ · 2026-09-28
- Harness-Zero: PKU, Google and HKUST Distill Agent Harnesses into Model Weights — AxSaucedo · 2026-09-28
- PromptSentry: open-source 3-layer proxy blocks prompt injections in under 1ms with local DLP scrubbing — Ok-Negotiation342 · 2026-09-28
- ProgramBench Called an Excellent, Far-from-Saturated Eval for Multi-Agent Coding Systems — jyangballin · 2026-09-28