Hadlock
@Hadlock · Sep 22
Jev-like Doom Bot proof of concept in python
Entries
ChristianAlexander
@ChristianAlexander · Sep 20
An Elixir library wrapping ConvAI Innovations' OSS Laya system one model
aovestdipaperino
@aovestdipaperino · Sep 20
Pure-Rust inference for the Laya non-autoregressive typed-decision model (ModernBERT-large + RL decision head), on candle
glukicov
@glukicov · Sep 20
Model routing is a System 1 job: a 421M local decision engine (Laya) against a GPT-5 nano reasoning router. Same accuracy, 35x faster, free.
yibie
@yibie · Sep 20
Independent measurements of typed-decision models: Jev (TypeSafe API) vs Laya (open weights), and a local-first cascade that matches Jev's accuracy at 1.8x the speed
allenporter
@allenporter · Sep 20
Conversation agent based on Laya, a multilingual, non-autoregressive System 1 decision model.
dhruvmehra
@dhruvmehra · Sep 22
Benchmark TypeSafe JEV against LLMs, fine-tuned BERT, Laya and zero-shot NLI on text classification: accuracy, calibration, latency, throughput, cost
ZLHAOOO
@ZLHAOOO · Sep 25
Chinese fine-tuned Laya decision model weights (MLX, Apple Silicon) — 中文微调 Laya 决策模型: 消息路由/紧急度/优先级/检索选词, 一次前向全出, ~27ms
cv
@cv · Sep 20
World 1-1 cleared with unmodified Laya on DGX Spark. Model-driven jumping, replay-verified frames, and every decision recorded.
chneau
@chneau · Sep 20
Dockerized FastAPI service for Laya typed-decision predictions: multi-checkpoint routing, API-key/Basic auth, presets and bulk inference.
aovestdipaperino
@aovestdipaperino · Sep 20
Browser pong whose paddle is decided by a Laya typed decision, one question per frame at 18.7 ms
myxamediyar
@myxamediyar · Sep 19
Long-input harness for Laya, the open-weights System 1 decision model: chunk, detect, aggregate. Apache 2.0.
wangmiaozero
@wangmiaozero · Sep 21
Cross-platform Laya decision routing skill for Codex, Claude Code, OpenCode, Pi and other Agent Skills compatible agents.
thomasbrueggemann
@thomasbrueggemann · Sep 19
A coding agent CLI where Jev (TypeSafe System One) or Laya decide what to do next and a configurable LLM does the work.
Adkid-Zephyr
@Adkid-Zephyr · Sep 21
Feishu message classification benchmark: 64 synthetic scenarios, reusable classifier adapters, and audited Jev vs Laya results.
cdepuy
@cdepuy · Sep 19
Hermes plugin: Jev-style skill pre-routing with a local Laya model (421M System-1). Injects the top-N relevant skill before the main model sees a task.
raboija
@raboija · Sep 22
Local Laya 322M typed decisions in a Pac-Man-inspired Windows app. NVIDIA CUDA, live probabilities, transparent planner guidance and reproducible benchmarks.
0x440-1me
@0x440-1me · Sep 22
Unity-native SDK for Laya and System One decision models — Choice, Noul, Score, local Laya, and TypeSafe-compatible backends.
Nandakishorm1
@Nandakishorm1 · Sep 22
We ran two decision models, Jev and the open-weights Laya, through one harness where only the engine changes. Jev led by 4.7 points alone and 6.8 with our feedback layer, and then fine-tuning Laya on the same 140 labels beat both.
lzero07
@lzero07 · Sep 22
Independent Chinese skill-routing eval of laya (System-1 decision model): 3 checkpoints × 20 real-world requests — accuracy, latency, confidence calibration. Windows/Py3.14 segfault workaround included.
