Builds made with Laya

介绍比Jev快50倍，在你设备上跑的laya-mlx!

只在你的设备上占用最高1G内存

Laya是一个开源的类似于Jev的,基于文本输出概率的分类系统

我将其移植到MLX,并且做了一些性能优化!

视频中就是这个模型在我的本地M3Max上玩贪吃蛇

这个模型能够以每秒决策60次的速度玩贪吃蛇！

Local Laya moggs Jev at @grok 4.7-built Tetris 🧩

An open-weights System One model called Laya, beat cloud-based Jev at playing Tetris by making decisions 11 times faster, running locally on a 16GB MacBook Air!

Run AI models locally ->

i made Jev and Laya play the chrome dinosaur game against each other

except they’re actually controlling the dinosaurs.

Laya runs locally on my Mac, Jev runs over an API.

same track, same physics, completely different latency.

they make their own decisions, dodge obstacles…

Thanks for the model, we were able to port Laya to coreml with 99.5% of the ops on ANE + benchmarked too. it is now blazing fast with 3.7 ms per decision on an M5 Pro.

Release: https://t.co/9xerYIHt9y

Models:

Tetris bench: Jev vs. Laya-mlx

I run the Laya model on my MacBook. It's true that the model is really fast (~84ms), but it's also much more stupid.

Losing to Jev 3 out of 3 rounds.

The open-source local version of Laya crushes Jev in response speed!

On the left is Laya, a local 421M open-source decision model.

On the right is Jev 1.13.0, running through a cloud API.

Same Snake game, same typed decisions, and the same 30-second free run:

• Laya: Score…

If you’re wondering how fast Laya is compared to Jev:

my maze demo measured 19ms median responses from local Laya on a base M5, versus roughly 300ms from Jev’s cloud API.

Nerve is a supervisory nervous system for Hermes agents, adding typed System One decisions, ranking, verification, token-aware oversight, and an opt-in tool gate powered by TypeSafe Jev or Open Source Laya

The problem isn't the models, it's how you define the problem space, design the movements / decisions and optimize for it (batching e.g.)

Jev wins for a network-based model any day.
Local models tied - Laya-mlx and my custom build Jax.

Look at them go.

Using Laya wasm for workflow automation

Jev vs locally running Laya-MLX and Kev-4B

I built my own snake benchmark. Jev @typesafeai is running via API.
The other models are small alternatives running on very little RAM on my Macbook!

Jev seems to deliver the best quality no doubt! After running it for a while 0…

Seven open source Jev models tested in 21 minutes. Laya gets the last 3:41

0% - 0:00 - intro
17% - 3:46 - JevBench, the thing they all get scored on
33% - 7:02 - Nimble
51% - 11:03 - Decider
62% - 13:20 - OpenJev
83% - 17:50 - Laya

The order is the story. Laya did 2M views…

Voice-control Safari and Chromium browsers on Apple silicon with a local Laya decision model that checks each result.

Update: I trained Laya non-step for close to 24 hours. No where close to jev 0-shot :|

جالب شد بازی اسنیک. ورودی لوکیشن میوه و هد و بوردر هارو میگیره و با مدل اوپن سورس #Laya به صورت لوکال که شبیه #Jev هست دستور میگیره که لوکیشن بعدی کجا بره بالا پایین چپ راست.

تجربه جالبی بود

اینم لینک پروژه تو گیت هاب ممنون میشم با 🌟 حمایت بکنین:

Local browser AI agent with indexed action space — Laya local decisions, /opt/homebrew/bin/bash/run. Google Flights demo in ~6.5s. Python.

Paper trading with Laya MLX on crypto (Binance) and S&P 500 stocks (Yahoo Finance): live dashboard, backtests, prompt lab and rolling walk-forward tests, all local on Apple Silicon

i made Jev and Laya fight 1v1 each other in Doom

1v1 deathmatch, monsters in between, first to kill the other opponent 5 times wins!

@typesafeai Jev's calls were sharper, it needed a third of the corrections.
@mizorewww Laya just decided twice as often.

final results
Laya…

Wow, Laya (https://t.co/PhmZAwTZeX)
inference in Jolt (Clojure running on Chez Scheme)

Experimental autonomous RimWorld colony management powered locally by the open-weight Laya decision model.

完全离线的 Android 决策小工具：用本地的 Laya 决策模型给选项排序（ONNX Runtime + int4 量化）

Porting the laya-mlx to dgx spark GB10

Beating Laya (https://t.co/rTxWDMS8KG) in tic tac toe : )

Using this for Unity https://t.co/XEWtg1ZaEd by @asus4

Analyse any text or ebook on 82 calibrated dimensions (genre, mood, themes, style) with the Laya model — ~1 s per page on Strix Halo with ROCm 10. Model: huggingface.co/convaiinnovations/laya