Research with Laya

Video by AI WITH Rithesh covering Laya: Laya Is INSANE… Open-Source AI Just Challenged Jev!. Part of the early wave of tutorials and head-to-head tests after the open release.

Video by Nick Automates covering Laya: Laya is a Free and Opensource alternative to Jev AI but faster. Part of the early wave of tutorials and head-to-head tests after the open release.

Video by Tonbi's AI Garage covering Laya: Laya: Fine-tuning the Open Jev-style Decision Model on Skill Routing. Part of the early wave of tutorials and head-to-head tests after the open release.

Video by Singularity Feed | AI Models , Agents & Robotics covering Laya: I Tested Laya: An Open-Source JEV Alternative. Part of the early wave of tutorials and head-to-head tests after the open release.

Video by Data Science in your pocket covering Laya: Laya : Bye TypeSafe Jev AI . Part of the early wave of tutorials and head-to-head tests after the open release.

rbrus

@rbrus · Sep 22

Tests whether the open Laya model on Apple's MLX can stand in for a paid judge, scoring model outputs locally with typed answers.

glukicov

@glukicov · Sep 20

Two LLM routers behind one endpoint compared on 180 labelled requests: Laya picks small, medium or powerful in one forward pass versus a GPT-5 nano reasoning router. Local and free for the Laya side.

yibie

@yibie · Sep 20

Independent Jev-vs-Laya measurements plus a local-first cascade: Laya decides first, high-confidence answers execute, low-confidence ones escalate to Jev. Reported as matching Jev's accuracy at lower cost.

yankis

@yankis0x · Sep 21

Yankis's video runs Nimble, Decider, OpenJev, Laya and more through the same JevBench scoring. Laya gets the final 3:41.

biplovgautam

@biplovgautam · Sep 22

A studio for fine-tuning Laya typed-decision models locally, with checks that prove the tuned model beats the baseline before you ship it.

Dima Nurm

@NURM_Dima · Sep 21

Dima Nurm sorted 1,000 emails with the 322M Laya model on an M3 Air: 28.6 s without screen recording, 65.1% accuracy against reference labels, 0 MB swap, no cloud.

antonellof

@antonellof · Sep 24

Compares Laya MLX's typed move choices against Dijkstra on seeded weighted mazes, with a side-by-side replay on Apple silicon.

ameeetgaikwad

@ameeetgaikwad · Sep 24

A real-time traffic simulation where on-device Laya and cloud Jev control identical cities so you can compare their decisions.

Adkid-Zephyr

@Adkid-Zephyr · Sep 21

Synthetic Chinese workplace-chat scenarios with frozen inputs, prompts and labels. Reports classification quality, mis-generated tasks and latency: Jev 253 ms vs Laya 151 ms.

A post pointing to a third-party Jev-vs-Laya study as evidence that task-specific fine-tuning closes, and often flips, the accuracy gap between the open Laya model and the closed Jev.

DDnim

@DDnim · Sep 21

Fan-out SUMs, = NULL, UPDATE without WHERE, injection remnants and more, each with intent and schema context. Four typed questions per statement, identical payloads to both models.

bvolpato

@bvolpato · Sep 21

One import, no server. Dependency-free Rust compiled to Wasm with WebGPU kernels; pulls a pinned int8 pack from Hugging Face and keeps it in the browser. Live Tetris included.

Alexander-Ollman

@Alexander-Ollman · Sep 21

Matched-budget fine-tunes on Banking77 and five more public workflows (prompt injection, SMS spam, emotion, request routing), plus two Aegis chat-moderation fine-tunes.

snsk

@snsk · Sep 21

Automotive, embedded, enterprise software and product decisions, three trials each, with questions, expected answers, raw responses and a static HTML report.

mouadse

@mouadse · Sep 21

Reproducible zero-shot benchmark on informal Moroccan Darija reviews in Arabic script and Arabizi, positive/neutral/negative. Deterministic seed-42 split, no training.

elcronos

@elcronos · Sep 20

A frozen-protocol comparison of Jev 1.13, PrismNLI-0.4B and Laya with raw predictions, calibration and latency published. Laya ties Jev on the emotion dataset and trails it by 10 to 33 points elsewhere.

umstek

@umstek · Sep 20

Seven zero-shot extraction and classification families demoed, benchmarked and cross-compared, with sizes, licences and costs in one table.