World Summit Tournament: a fighting game with a Laya opponent
Daniel F's browser fighter has 21 characters, special moves, combos and replays, with the computer opponent run by Laya.
BUILDS
113 builds made with System One models.
Agents, routers, scorers and games — built on models that decide. Every entry links to the models it uses, so you can pull the same one and start from there.
113 builds · page 4 of 5
arbiter: a Jev-compatible decision server for Laya or your own model
A self-hosted typed-decision server for NVIDIA GPUs and Apple Silicon that speaks Jev's wire format, with integrations for coding agents.
docker-laya: a complete Laya API service in one container
A FastAPI service that bundles all three Laya checkpoints behind a language router, with API keys, presets, OpenAPI docs and bulk inference.
EdgeJev: offline Laya on a 4-core CPU in 15.6 ms
A convert, quantize and deploy toolchain that turns Laya and related models into INT8 ONNX for offline CPUs, measured at 15.6 ms per question.
A fully local Home Assistant conversation agent on Laya
Allen Porter's Assist agent resolves intents and entities in a single Laya pass on CPU, CUDA or Apple Silicon, with nothing sent to the cloud.
Reranking Finnish legal passages with typed-judgment models
Compares Jev, open-weight Laya, cross-encoders and GPT models as rerankers over a Postgres full-text and pgvector first stage, with cost per 1,000 queries.
jevals: agent evals and guardrails as typed decisions
Openlayer's library bundles the evaluations for an agent trace into one decision call, cheap enough to run on every trace — or free with local Laya.
laya-candle: Rust inference with custom Metal kernels
Runs the original Laya safetensors through Candle with fused kernels, tiled attention and GEMM dispatch tuned for the M1 Pro.
laya.cpp: a standalone C++ runtime with CUDA and Vulkan
A ggml-based runtime for all three checkpoints with native tokenization, decision heads and JSON output.
laya_ex: Laya in Elixir on Nx and Bumblebee
Fetches the official checkpoint on first use and answers typed questions natively, with EMLX on Apple Silicon.
Laya checkpoints in GGUF
GGUF conversions of the English and multilingual models for llama.cpp-style local runtimes.
laya-jev-lab: Laya first, Jev only when unsure
A local-first cascade: confident Laya answers run straight away and uncertain ones escalate to Jev, matching Jev's accuracy at 1.8× the speed.
laya-mcp: laya-mlx for any MCP client
Will Sargent's stdio MCP server lets Claude Code and other clients classify text, score against a rubric or ask yes/no questions locally.
laya-mlx plays Snake at 60 decisions a second
The announcement of the MLX port: under 1 GB of RAM, choosing the snake's next move sixty times a second.
Laya clears Super Mario Bros. 3, World 1-1
Unmodified weights on a DGX Spark, real NES emulation and text observations read from RAM: 207 verified decisions at a 16.79 ms median.
Laya plays Tetris, Flappy Bird and a lane runner on a laptop
A live site from brain function collapse where the model plays on its own at roughly 27 ms per decision.
laya-pong: a browser paddle at 18.7 ms per frame
One typed question per frame on Metal, with the opposing paddle driven by three lines of arithmetic for comparison.
laya_router: as accurate as GPT-5 nano at routing, 35× faster
Two routers behind one endpoint on 180 labeled requests; Laya picks small, medium or powerful in a single forward pass, for free.
laya-rust: pure-Rust Laya inference
Loads the safetensors directly with candle and runs any checkpoint on CPU, Metal or CUDA — no Python, torch or ONNX export.
Fine-tuning Laya for real-time Tetris placements
The 421M model learns to choose among enumerated placements scored by a board-evaluation teacher, with a guide to reproduce it.
Laya Warehouse Safety: a robot that knows when to wait
In a replayable grid with moving workers and forklifts, Laya chooses advance, shift or wait and flags collision risk, under a deterministic safety shield.
laya-web: the whole model in a browser tab
524 MB of 8-bit English weights cached in the tab and run with onnxruntime-web, live at laya-web.pages.dev.
@r4ai/laya-web: client-side Laya with WebGPU
A browser runtime on ONNX Runtime Web that uses WebGPU with a SIMD Wasm fallback and works inside Web Workers.
LayaKit: one Core ML decision per call from Swift
A Swift package derived from laya-coreml that loads the general or Neural Engine bundles and predicts in a single call.
Many of the first builds here were collected by madewithlaya.com. Every entry links to its creator’s own post, repository or site.