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
decision-systems-bakeoff
actuallyrizzn
Runs a reproducible benchmark comparing Flybrain, Jev, and Laya on four public tests
laya-studio
Hantlowt
Specializes frozen Laya representations for new decision tasks and exports portable artifacts
traffic-solver-laya-
alipyth
Simulates city traffic controlled by Laya and includes a model playground and tutorial
vegaml
NandhaKishorM
Produces calibrated typed decisions from frozen language models with conformal sets and abstention
chunklaya
myxamediyar
Chunks long documents and uses Laya for typed-decision lookup across inputs up to one million tokens
zero-shot-ie-bench
umstek
Compares Laya and other zero-shot systems across information-extraction and classification tasks
structured-decision-bench
zhengbangbo
Benchmarks structured decisions from Qwen3, Jev, and Laya CoreML
doomLaya
azalio
Trains and compares Laya and Jev agents playing FreeDoom with reproducible results
Laya-Showcase
zamax14
Demonstrates and benchmarks multilingual Laya alongside Jev and GPT models
laya-ternary-lite
xixi3548942758-design
Quantizes Laya to ternary weights for smaller, lower-memory inference
flappy-laya-openvino-cpu
rupeshs
Runs Laya inference with OpenVINO to control a Flappy Bird game on CPU
laya-thalamus
lcbkmm
Middleware uses Laya for agent tool routing with confidence-based LLM fallback and evaluation
open-jev-laya-bench
pngwn
Shared benchmark of Laya 421M vs System One 4B and Bespoke Nimble 9B on accuracy and calibration
laya-multilingual-FP8
Weidows
Provides an FP8 per-row quantization of the multilingual Laya checkpoint with multilingual evaluation results
laya-ov-int8
rupeshs
An int8 OpenVINO export of Laya for CPU inference
laya-turkish-mmlu
AhmetSemih
A multilingual Laya checkpoint fine-tuned on Turkish MMLU questions
laya-multilingual-onnx
Emerald7664
CPU-optimized ONNX checkpoints for multilingual Laya decision inference
sokudan
hiroki-abe-58
Builds a Japanese typed-decision model and benchmarks it against Laya
LayaSnakeGame
AmirTahaMim
A Pygame Snake game compares human play, a greedy baseline, and local Laya decisions
laya-snake
zxrneu
Builds an autonomous Snake game agent using Laya decisions and BFS pathfinding
jev-vs-laya
DDnim
Benchmarks Jev and Laya on typed decisions for reviewing SQL statements
doom-war
rythmn1111
Compares local MLX Laya and hosted Jev as decision-makers in a Doom deathmatch
laya-vs-jev
zaferayan
Benchmarks multilingual Laya against hosted Jev across 900 cases, three tasks, and six languages
laya-cuda
Alexw1111
Provides a lightweight CUDA inference library for Laya
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