Jev directory
CommunityJev Models & Reimplementations
Spot Snake MiniMind-JEV v0.1
A MiniMind-JEV ranking head post-trained to pick moves in Spot's Biscuit Snake. The game offers the safe moves and the model ranks them, copying a teacher that keeps the most open room. Ships head.safetensors, a frozen BGE encoder, metrics files and a local serve script under MIT.
- Reading time
- 2 min
- Published by
- Community
- Added
- Nov 2026
Highlights
- Post-trained from cnfeichen/minimind-jev-v0.1 at revision fa26d3e20f to rank safe moves in Spot's Biscuit Snake.
- Selected checkpoint scored 55.80% teacher agreement on a 1,708-row held-out test.
- Best of 5 epochs reached 57.70% teacher agreement on 1,785 validation rows.
- Ships head.safetensors, a frozen BAAI/bge-small-zh-v1.5 encoder, and metrics.json, training.json and reload-check.json.
- No GGUF release: llama.cpp has no converter for the custom MiniMind ranking head.
Questions about this entry
Short answers drawn from the linked project or page.
It is a MiniMind-JEV ranking head post-trained to pick moves in Spot's Biscuit Snake. The game offers the safe moves and the model ranks them, copying a simple teacher that keeps the most open room. The head is a ranker, not a text generator.
Download it with hf download Lucasgitguy-co/spot-snake-minimind-jev-v0.1 --local-dir models/spot-snake-minimind-jev-v0.1, then run python scripts/serve.py --model models/spot-snake-minimind-jev-v0.1 to serve the local game on port 8000. The scripts, dataset generator and INSTRUCTIONS.md are in the dataset repo.
It is how often the model picks a move the teacher accepts. It is not game score, win rate or speed; those have not been measured for this model.
Base model validation was 56.30% on 1,785 rows. The best of 5 epochs reached 57.70% on the same validation set. The selected checkpoint scored 55.80% on a 1,708-row held-out test, and reloaded saved files scored 56.25% on a 32-row test subset.
No. There is no GGUF release because llama.cpp has no converter for the custom MiniMind ranking head.
It ranks only the moves the game already marks as safe, so the game's collision checks must be kept. The teacher is a heuristic, not an optimal player, and matching it is the whole goal here. Confidence scores are not calibrated after training.
MIT covers the MiniMind-JEV head, the BGE encoder and the new Spot Snake training. Licence texts are in LICENSE, NOTICE and licenses/. The base model is cnfeichen/minimind-jev-v0.1 at revision fa26d3e20f.
Reactions & coverage
Posts, threads, and videos about this entry from around the web.
Hacker News: Show HN: OpenDecision, a local decision model with a Doom demo.
dwa35926 pointsShow HN: OpenDecision – a 400M zero-shot model makes local decisions, plays DoomRead the thread on Hacker NewsX: Standard One SH (StandardThinking)
LinkedIn: What multimodal decision models actually change, and what they do not.
LinkedInJevDirectory.org
Hacker News: Decision Index 0.2.1 scores open reproductions of Jev on a frozen suite.
jackbravo4 pointsDecision Index 0.2.1 – 70 open reproductions of TypeSafe Jev's Decision ModelRead the thread on Hacker NewsMore Jev resources like this
An Apache-2.0 decision engine with a TypeSafe-compatible endpoint, document answers with source passages, and a ViZDoom demo that picks actions from structured game state.
Jev Models & Reimplementations#community#python#local
Community
deepanwadhwa/OpenDecisionPython
deepanwadhwa/OpenDecisionPythonA native MLX runtime for Laya's open typed-decision weights: short decisions on Apple Silicon, with no generated tokens, PyTorch, or cloud API.
Jev Models & Reimplementations#community#python#open-models
Community
mizorewww/laya-mlxPython
mizorewww/laya-mlxPythonAn open-model semantic-if engine positioned as a local Jev-style decision stack, separate from the original SemIf repo but sharing the same author.
Jev Models & Reimplementations#community#python#open-models
Community
TheoLeeCJ/SemIf-OpenJevPython
TheoLeeCJ/SemIf-OpenJevPythonLaya typed decisions through Apple Core ML and the Neural Engine, with the author's speed and energy benchmarks for short decisions on an M3 Max.
Jev Models & Reimplementations#community#python#open-models
Communitymizorewww/laya-coremlPython
mizorewww/laya-coremlPythonA local yes/no decision model that speaks Jev's wire format on a laptop CPU, refusing Choice and Score until those question types ship.
Jev Models & Reimplementations#community#python#open-models
Community
feder-cr/jevC++
feder-cr/jevC++A local server that reads next-token probabilities from a fine-tuned Spark model and returns Choice, Score, and Noul answers with zero generated tokens.
Jev Models & Reimplementations#community#python#open-models
Community
Rizzo-AI-Academy/rizzo-flowPython
Rizzo-AI-Academy/rizzo-flowPythonFrom the Jev guides
Original write-ups on what this entry is for.
From the Jev AI model community
Posts from builders shipping with Jev right now.