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.
468 projects
laya-rlcd-benchmark
mourad-ghafiri
Evaluates the multilingual Laya model on adversarial decision-making scenarios
jev-arena
Samge0
Compares Jev, NanoJev, and Laya decisions in Tetris and 1024 with recorded gameplay replays
computer-use-testing
yoaugust
Tests Jev, Laya, and CUA for computer-use tasks and plans further reinforcement-learning fine-tuning
laya
DarkWanderer
Runs experiments with the open-source Laya System-1 decision model
laya
ahmed-alqasaby
Evaluation grid and Kaggle training notebooks for Laya checkpoints
jev-laya
wuzhiping
Probes the multilingual Laya model on Apple Silicon and records machine-readable inference results
laya-demo
tjpajala
Benchmarks Laya and Open-Jev on JevBench and PubMedQA with a Dockerized comparison interface
laya-ft
Alexander-Ollman
A reproducible study benchmarks Jev against fine-tuned Laya on workflows and chat moderation
laya-doom
cohenom
Displays Laya's live tactical votes alongside heuristic-driven Doom combat
laya-chess
buzzo123
Provides chess games against a computer that selects moves using Laya
Laya-test
LouisMoretti
Runs Laya locally for message classification and benchmarks its server and prediction performance
laya-mlx
MohammadAsadi-7
Runs Laya typed-decision models natively on Apple Silicon using MLX
laya-demo
cyyeh
Provides a Gradio playground, CLI scenarios, and device benchmarks for Laya
laya-test
petrixh
Runs Laya as the decision-making pilot in a browser game with reproducible evaluations
Laya-2048
DjTaNg-404
Uses local Laya weights to play 2048 with browser, terminal, and batch-testing interfaces
laya-snake-cuda
kuchris
Compares Laya and SemIf for local real-time Snake decisions with a CUDA dashboard
lunar-mpc-laya
mraad
Pairs Laya with adaptive model-predictive control in a lunar lander decision game
laya-mlx-demo
knishika62
Uses laya-mlx to classify Japanese virtual streamer personas and compare results with an LLM
jev-vs-laya
mouadse
Benchmarks hosted Jev, Kev-9B, and Laya on Moroccan Darija sentiment classification
laya-onnx-runtime
zenryokukikai
Ports the Laya multilingual decision model to CPU-only ONNX Runtime
laya-snake-arena
sathwikkuncham
Compares local Laya and Jev decision engines in a configurable Snake arena
jev-laya-chess-bench
abe17124
Compares TypeSafe Jev and Laya in head-to-head chess games using legal moves
jev-laya-japanese-business-benchmark
snsk
Compares Jev and Laya on a Japanese business decision benchmark
decisionmakertest
vittoriobrehautduran
Benchmarks Jev and Laya on typed decisions, labeled cases, and a small dungeon game
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