Automated README summary
Automatically summarized from the README on 13 Oct 2026; not a run by jevbooks. Run instructions and author-reported outcomes still need verification.
What it is
Valen is a multimodal decision model inspired by Jev. It evaluates text, images and video against task instructions and returns probabilities over supplied candidates, giving software a structured decision interface. The repository includes model implementation, data processing, SFT and RLCD training, and inference/evaluation commands.
How it uses Jev
Valen uses Jev-like typed questions (Choice, Score, Noul) on a state that can include text, images and video. The model returns calibrated probabilities over candidates. In code, call model.predict with a state and questions dict, using execution='shared_state' or 'question'. Results are used for decision-making, evaluation, and game rollouts.
Primitives:choicescorenoul
Technique worth stealing
Shared-state inference with Mixer, trained via two-stage SFT and RLCD.
Try it
Run cookbook notebooks: API quick start or local inference. Install dependencies, then use model.predict.
Model judgments from README excerpts
Project signals use repository metadata, an existing summary and the first 700 README characters, not the full project. “Not found” means not detected in that material; it does not establish absence. These are the original judgments with recorded models and dates, not source audits or run results.
Models & evaluation dates
- Inclusion gate
jev-1.13.0- Project signals
jev-1.13.0- Category
jev-1.13.0- Pattern tags
jev-1.13.0
Evidence
These signals use repository metadata, an existing summary and at most 700 cleaned README characters. ≥ 0.60 means reported in that excerpt; ≤ 0.40 means not found there. The middle band is inconclusive. None is a source-code audit or a test result. The runnable-tool question excludes research experiments; the recommendation question asks for a learning shortlist, not a quality verdict.
- Jev is core to the projectFound in material0.70
- Shows a System One patternnot found in the excerpt0.15
- Describes handling uncertaintynot found in the excerpt0.04
- Reports measured numbersnot found in the excerpt0.10
- Is a runnable toolunclear0.42
- Top-10 learning shortlist signalnot found in the excerpt0.33
- Is a replica of the modelFound in material0.96
- Solves a common problem (0–2)score on a 0–2 scale1.25 / 2
- About JevFound in material0.90
no-judgment band (0.40–0.60)
no-judgment band (0.40–0.60)
The Jev model and date for each judgment stage are listed above. Card written by DeepSeek V4.1 Flash from the README on 13 Oct 2026.