Laya alternatives
Jev was released in September 2026, and within weeks several open models with the same typed-question interface appeared. Laya is a 322M to 421M encoder. Most alternatives are open LLMs, mostly Qwen, fine-tuned at 0.5B to 35B parameters to score options without generating text. They trade speed for accuracy.
472 projects
laya-onnx
D-Ajay-Kumar
Converts Laya weights to ONNX and benchmarks the converted model
laya-chess
Srimanvarma001
A local chess app compares Laya-suggested moves with a one-ply material check
laya-local-setup
jigs10
Provides commands and examples for running Laya locally and comparing its decisions with Jev
laya-client-example
angelxmoreno
Uses a Laya HTTP client to triage GitHub issues and compare category predictions with repository labels
jev-vs-laya
soummyaanon
A web-based arena compares Jev and Laya classifiers on shared tasks and reports performance metrics
laya_demo_test
j75689
A Snake demo compares Python and Rust Laya runners with ONNX Runtime and Core ML backends
app-intents-laya-study
junshanxu
Studies macOS App Intents and CLI agents with and without a local Laya next-action gate
jev-laya-cube-arena
189-sketch
Compares Jev and local Laya on Rubik’s cube next-move decisions
laya-alcohol-visit-classifier
Aurumdev952
Fine-tunes and benchmarks Laya for alcohol attribution in hospital visit notes
pi-context-gc
Lausi-h2ai
Archives old tool outputs for Pi coding agents and evaluates retention policies with local judges
jev-decomposition-tr
betulsimsek
Tests question decomposition for typed decisions by comparing Jev, Laya, and a local decoder
impulse
roby-avo
Creates a physics arena to play against decision models and inspect their choices
system1-judge-eval
wanqili857-byte
A reproducible evaluation of Laya and other System-1 models on out-of-domain agent decisions
system-one-playground
goodboybeau
A macOS playground comparing Laya, Decider, Kev and Jev on accuracy, calibration and performance
layaUI-HK
whksoup
Runs a local Laya playground, games, benchmark, and agent skill
jev-spaceshooter-demo-laya
jstdlee
Steers a retro space shooter with Laya decision models through a local API
layarerank
anguyenbus
Reranks documents in batches with Laya decision models and evaluates fine-tuned checkpoints
laya-eval
imaddde867
An evaluation compares local Laya with jevmlx and hosted Jev on public cases
laya_test
liyan24
A Chinese-language test suite evaluates Laya and Jev with a 120-question set and dual-backend comparison
laya-escalate
Sidharth-06
A Python package audits Laya confidence calibration and provides repairs and escalation policies
Laya-lab
DamianPrist
Compares local Laya with DeepSeek and tests Laya's choice-position effects
laya-lambda-cdk
KICU1129
Deploys Laya inference on AWS Lambda and reports Japanese classification and latency evaluations
jev-laya-trader
PrateekReddy116
Compares Jev and Laya decision engines in a paper-trading sandbox
jev-laya-benchmarks
stemirkhan
Benchmarks Jev and Laya against specialized rerankers on BEIR datasets
The models
Sizes and licences are from each model's Hugging Face card.
| model | maker | sizes | licence |
|---|---|---|---|
| Kev | Jared Palmer | 0.5B to 27B (benchmarked: 0.8B, 4B, 9B) | Apache 2.0 |
| Decider | Mapika | 0.8B to 35B, plus a 2B vision model | Apache 2.0 |
| Clef, Clef-flash | Cloudflare | 27B, 9B | Apache 2.0 |
| GLiNER2.5-Decide | Fastino | 340M English, 287M multilingual, 1B | Apache 2.0 |
| Julia-1 | Supersonic Labs | 144M | Apache 2.0 |
| Jebadiah | Frontier Infra | 4B, 9B, 27B | Apache 2.0 |
| pplx-decider | Perplexity | ~26B ("27b" in the name) | Apache 2.0 |
- Kev runs a frozen Qwen3.5 base prefill-only, with a LoRA adapter and a pointer head over the options. The card says states of up to 65,536 tokens are served and 8,192 are validated.
- Mapika Decider calls itself an open reproduction of the System One model class. Its training code and data registry are on GitHub.
- Clef and Clef-flash read text, JSON, images and video, per their cards. Cloudflare's announcement says they are Jev-API compatible and also served on Workers AI.
- GLiNER2.5-Decide is a classification-only member of the GLiNER2.5 family: any label set at call time, single- or multi-label.
- Julia-1 is, like Laya, an encoder: it starts from mmBERT-small. See ModernBERT and mmBERT in Laya.
- Jebadiah ships a standalone server that answers
POST /v1/systemoneon one CUDA GPU or an Apple Silicon Mac. - pplx-decider v1.1 uses a separate decision readout instead of a normal language-model head, so its card asks you to use the included inference code.
The hosted reference is Jev from TypeSafe. Its weights are closed and it is billed per input token. The differences from Laya are in Laya vs Jev.
What independent benchmarks found
akitaonrails (October 10, 2026) used synthetic, AI-assisted cases in English and Brazilian Portuguese without independent human labels. Chance was about 31%. Local models ran on a Ryzen AI Max+ 395. Compact-set accuracy and median latency:
| model | accuracy | latency |
|---|---|---|
| Jev (hosted) | 97.6% | 255 ms |
| Clef (hosted) | 95.8% | 302 ms |
| Perplexity Decider 1.1 | 99.4% | 1,547 ms |
| Jebadiah 4B v2 | 90.5% | 205 ms |
| Kev 9B | 88.1% | 561 ms |
| Kev 4B | 87.5% | 246 ms |
| Mapika Decider | 87.5% | 186 ms |
| Clef-flash (local) | 87.5% | 618 ms |
| Kev 0.8B | 63.1% | 86 ms |
| GLiNER2.5-Decide (English only) | 52.4% | 208 ms |
| Laya (English) | 42.9% | 41 ms |
| Julia-1 | 31.5% | 12 ms (GPU) |
The author's conclusion: if Jev-like accuracy is a requirement, Laya, Julia-1 and the 0.6B Decision 3.0 Edge are not usable out of the box. He also notes that judging these small encoders by the accuracy column alone misreads them: they are built to be fine-tuned and run at the edge. Fine-tuning was not tested.
Arize (October 2026) ran a bracket of hallucination detection on RAGTruth, LLM-evaluator suites and agent routing. Balanced accuracy on RAGTruth was 85.5% for Jev, 85.2% for Clef 27B, 76.8% for Kev 9B and 59.8% for laya-multilingual. Laya was the fastest model at 0.04 s p50 on an M4 Max, and had the highest calibration error (0.330).
systemonemodels.ai lists 29 open Jev alternatives with the accuracy each publisher reports. Those figures come from different test sets, so they cannot be compared with each other.
How to choose
- Speed, CPU or on-device: the encoders. Laya and Julia-1 answer in tens of milliseconds, and Laya has the most ports, from Mac to the browser. Plan to fine-tune or to keep a fallback for hard cases.
- Accuracy without training: the 4B and larger models. They need a GPU or an Apple Silicon Mac with enough memory, and they are several times slower.
- No servers: Jev, Clef on Workers AI, or a hosted Laya API.
Whichever you pick, run a few hundred of your own labelled examples through two or three candidates. The rankings above come from small or synthetic sets, and their authors say so.
More ways to use Laya