Resources

🚀ms-swift now supports LLaMA model training!!

#36 opened about 18 hours ago by

z0o0ey

laya-guard: Laya judging coding-agent shell commands before they run (live demo + field notes)

#35 opened 6 days ago by

Ayaannnnnnn

Йост и малютка

#34 opened 17 days ago by

nimirei

Vision graft on Laya: meal photo → calories/macros + food questions (LayaMacro, open source)

#33 opened 19 days ago by

Aquos06

laya-train 0.4.0 fails on cuda/mps before training starts: base checkpoint evaluated while the model is still on CPU

#32 opened 21 days ago by

MrSnuffy1972

Add decision-model tag

🚀 1

#31 opened 22 days ago by

victor

Seb-9B vs Jev 1.13 vs Laya on the 17 public suites from Laya's benchmark report (identical questions, self-run)

#30 opened 22 days ago by

ironbcc

🔒 [SECURITY ALERT] Alignment Calibrator & Robustness Benchmark Available

#29 opened 22 days ago by

urbisjuridica-ctrl

🔒 [SECURITY ALERT] Alignment Calibrator & Robustness Benchmark Available

#28 opened 23 days ago by

urbisjuridica-ctrl

very poor performance

#25 opened 24 days ago by

josie-ford57

Laya performance on fontlab.org/ornotto

#24 opened 28 days ago by

fontlab

Including models for router with higher input than 8192?

#23 opened 29 days ago by

anrichvdwalt

Update README.md

#22 opened about 1 month ago by

rahul05ranjan

Laya as a judge!

#21 opened about 1 month ago by

rbai86

Independent head-to-head on a sealed 1,190-case manifest: laya 0.686 vs Jev 1.13.0 0.907 vs Qwen2.5-1.5B PCD 0.605, with paired intervals, plus three adapter findings

🔥 2

#20 opened about 1 month ago by

saidutta69

Interest in acquiring Laya

🤯😔 12

#19 opened about 1 month ago by

dtemaat

cbjev: Laya checkpoints fine-tuned into a one-pass layout (6.7x faster on multi-question calls)

#18 opened about 1 month ago by

0010101010-1

Higher-Context

#17 opened about 1 month ago by

amarosnithe

Free hosted endpoint for Laya, Jev-compatible API

🚀🔥 3

#16 opened about 1 month ago by

oezi13

Zero-shot Laya as an agent step guard: agrees when agents say "done", never escalates destructive commands, blind past 320–768 tokens

👀 2

#15 opened about 1 month ago by

sinhal

Reproduced your 78.33% on MASSIVE exactly, but struggling on my own data. What am I missing?

#14 opened about 1 month ago by

JhouCode

A small real-world calibration check on Laya typed-decisions

#13 opened about 1 month ago by

NambiarONS

CPU latency numbers for laya-multilingual — the model card only has T4

#12 opened about 1 month ago by

htqmike

Real view of convaiinnovations/laya

#11 opened about 1 month ago by

RealFalconsAI

Decision model benchmarks

#10 opened about 1 month ago by

jglowa

Catalog selection: low gold-label accuracy and candidate-order sensitivity with Laya-MLX (multilingual included)

#9 opened about 1 month ago by

Ela7mady

First-run evaluation: K-12 school email triage (what worked, what didn't)

❤️ 3

#8 opened about 1 month ago by

nccyber

Laya feels like an embedding/similarity matcher, not a context-understanding model (benchmark included)

#7 opened about 1 month ago by

qwq95195

Independent Laya vs Jev head-to-head on identical inputs

👍🔥 3

#6 opened about 1 month ago by

saidutta69

Laya-MLX Metal Memory Growth Fix

#5 opened about 1 month ago by

finche

What was the dataset used?

👍 3

#4 opened about 1 month ago by

encryptedoreo

Laya-multilingual: a 100-scenario decision-making evaluation

🔥 19

#3 opened about 1 month ago by

mghafiri

choice:11+ temperature (0.1006) saturates confidence: answers return at 1.00, wrong ones included

👍 3

#2 opened about 1 month ago by

DrowsyJungle