# Laya Descent v4 decision model by Eren Terakye

> Laya Descent v4 selects lunar-lander manoeuvres from text game states. Read its released checkpoints, simulator results and evaluation overlap.

Updated 2026-11-12 (Australia/Adelaide)

## Quick answers

### What is Laya Descent v4 by Eren Terakye?

[Laya Descent v4](<https://decisionmodels.ai/models/laya-descent>): A [Laya](<https://decisionmodels.ai/models/laya>) fine-tune that copies a scripted lunar-lander autopilot for Low Gate.

[Laya Descent v4 facts](<https://decisionmodels.ai/models/laya-descent>)

[Source](<https://huggingface.co/eren8392/laya-descent-v4>)

### Who makes Laya Descent v4?

[Laya Descent v4](<https://decisionmodels.ai/models/laya-descent>): Eren Terakye

[Laya Descent v4 facts](<https://decisionmodels.ai/models/laya-descent>)

[Source](<https://huggingface.co/eren8392/laya-descent-v4>)

By Eren Terakye · Hugging Face weight release: 2026-11-08 · OPEN

A [Laya](<https://decisionmodels.ai/models/laya>) fine-tune that copies a scripted lunar-lander autopilot for Low Gate. The package supplies PyTorch weights, the MLX conversion used in the benchmark, training data and hashes. It is an artifact for one simulator and prompt, with manoeuvres selected from a fixed menu.

## Facts

- Base: convaiinnovations/laya at revision 7b928d828b7b0e022f929d9bd2e44165aa270148
- Inputs: Text game states and manoeuvre questions
- Licence: Apache-2.0; base licence and modification notice included
- Access: PyTorch and MLX checkpoints, data and training report

## Claims and sources

- **VENDOR-REPORTED**: The maker reports 142 landings out of 160 levels on courses 5001-5020, excluded from training and tuning. The training report separately contains 2 overlapping items among 2,728 evaluation rows. PyTorch and MLX choose the same manoeuvre on 5,795 of 5,801 recorded states; this open-loop parity check does not fly new courses. [source](<https://huggingface.co/eren8392/laya-descent-v4>)

## Links

- [Source: Card and checkpoints](<https://huggingface.co/eren8392/laya-descent-v4>)
- [Source: Training report](<https://huggingface.co/eren8392/laya-descent-v4/blob/main/torch/train_report.json>)
- [Source: Low Gate benchmark](<https://github.com/ErenTerakye/lowgate>)

## Related projects

No related projects are listed in the registry.
