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
exp-laya-router
zheyar-ltd
Demonstrates a CUDA-based Laya policy router in real-time Snake and Tetris games
laya-capability-business
matrix-air
Presents experiments measuring Laya’s capabilities, limitations, fine-tuning, and runtime migration
laya-vs-jev
darrenli6
Compares Laya and Jev in snake-racing and fighting-game arenas
jev-vs-laya
janagarajsn
A Flask app benchmarks Jev and local Laya on synthetic customer-support tickets
laya-jev-eval
yuvrajrox
Evaluates Laya against Jev on email intent and explores text generation with Laya’s backbone
laya-open-eval
lgy1027
Packages local Laya inference and evaluation with an optional HTTP adapter
laya-vs-llms
Ujjwal3115
Benchmarks Laya against hosted language models on developer commit triage and CI safety decisions
laya-zh-eval
lzero07
Evaluates Laya checkpoints on Chinese skill-routing requests
laya-support-ticket-triage
IshaanLabs
A local Laya system triages customer-support tickets and benchmarks results in an interactive dashboard
laya-or-jev-pong
bananadonn
Compares local Laya and hosted Jev decision APIs by having them control paddles in Pong
llm-jev-laya-bench
PerryLink
Measures the cost, latency, and failure boundaries of judgment layers
laya-gtx1650-benchmark
PredictiveManish
Measures Laya inference latency on a laptop with a GTX 1650 GPU
laya-vs-llm-benchmark
Shray15
Benchmarks Laya against a local LLM for selecting MCP tools
pol2-laya-studio
shentonyan
A local interactive playground visualizes and evaluates typed decisions from Laya models
ai-update-radar-lab
starhunt
Tests Laya-based project relevance decisions and compares results with saved Jev evaluations
Jev
Caho1
A research repository documents Laya fine-tuning, evaluation, local inference and browser automation experiments
trust-router
NotoriousPOG
Compares rules, BERT, and Laya for shadow-mode AI security routing
jev-frontend-lab
dante01yoon
Six frontend demos compare Jev and local Laya decision pipelines with optional LLM refinement
featherbench2
ed-is-ai
Benchmarks language models and typed-decision models on custom tasks and agentic decision sequences
yks-bench
UgurcanAkkok
Benchmarks local Laya checkpoints and hosted Jev on Turkish university entrance exam questions
PacmanLocalJev
raboija
Runs local Laya decisions in a Windows Pac-Man-inspired game with live probabilities and benchmarks
jev-demo
nadeem4
A demo site compares Jev and Laya playing games and includes benchmarks and agent experiments
faceid-bench
mazenDDr
A face-unlock benchmark measures detection and recognition pipelines and tests Laya for yes/no decisions
sys1bench
rssr25
A benchmark measures calibration, robustness, and other properties of typed decision models including 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