by Heman10x-NGU · AI 工具 · ★ 295
openJev-verdict-2.0: Non-Autoregressive System 1 Decision Engine -brightgreen) -success) openJev-verdict-2.0 is an open-source, non-autoregressive foundational decision model engineered for deterministic software automation, inspired by TypeSafe AI's Jev and Reinforcement Learning for Calibrated Decisions (RLCD).
| Stars | 295 |
| Forks | 47 |
| Language | Python |
| Category | AI 工具 |
| Quality Score | 59.9815173442794/100 |
| Open Issues | 3 |
| Last Updated | 2026-09-20 |
| Created | 2026-09-19 |
| Platforms | browser, python |
| Est. Tokens | ~15k |
These tools work well together with openJev-verdict-2.0 for enhanced workflows:
Explore other popular ai 工具 tools:
openJev-verdict-2.0 is Calibrated 151M Non-Autoregressive Decision Engine beating TypeSafe Jev & Laya on LocalLLaMA/typed-decisions (77.10% acc, 0.0636 Brier, 0.0144 ECE). It is categorized as a AI 工具 with 295 GitHub stars.
openJev-verdict-2.0 is primarily written in Python. It covers topics such as agentic-ai, brier-score, calibration.
You can find installation instructions and usage details in the openJev-verdict-2.0 GitHub repository at github.com/Heman10x-NGU/openJev-verdict-2.0. The project has 295 stars and 47 forks, indicating an active community.