Decision modelOpen source
NanoJev
TianyuCodings
- Released
- -
- Data date
- September 24, 2026
- Model class
- Access
NanoJev is a 0.6-billion-parameter model from TianyuCodings that brings the interface idea of Jev into a small local game and control model. It takes a game state, a question, and candidates, then returns Choice, Boolean, or Score as direct probabilities. The computation runs in parallel without decoding answer tokens.
The released unified-games-v1 variant was trained for Maze, Snake, ViZDoom, and Predict Position, so it remains focused on those four game and control tasks. The included source code is MIT-licensed; the Qwen base retains its upstream license.
Specifications and access
| Specification | Value and source |
|---|---|
| Model class | Parallel Jev-like decision modelSource |
| Access | LocalSource |
| Model ID | C-Tianyu/NanoJev@unified-games-v1Source |
| Input | Game states, questions and candidatesSource |
| Output | Structured probabilities, no text generationSource |
| Result | Choice, Boolean and Score as direct probabilitiesSource |
| Parameters | 0.6 billionSource |
| License | MITSource |
| Scope | Game and control tasks covering Maze, Snake and ViZDoom.Source |
| Notes | The released variant is trained on the documented game tasks and is not a general classifier.Source |
| Base model | Qwen3-0.6BSource |
| Choice options | 2-255 per questionSource |
Popularity
42.2 / 100
Sources and data date
| Type | Evidence and data date |
|---|---|
| Additional source | NanoJev repository (retrieved September 24, 2026) · NanoJev repository · Editorial description reviewed October 2, 2026 |
| Additional source | NanoJev checkpoint (retrieved September 24, 2026) · NanoJev checkpoint · Editorial description reviewed October 2, 2026 |
| Additional source | OpenRouter: Popularity index (derived), CC BY 4.0 |
| Additional source | Vercel AI Gateway: Popularity index (derived), CC BY 4.0 |
| Additional source | Creative Commons Attribution 4.0 |