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
Jev-OpenSource-Alternative-DEMO-Laya
Nagwa-Ahmed
Evaluates local Laya classification across Hugging Face datasets and exports Excel reports
aubin
Emrevrg
Provides calibrated decision models, computer-use tools, and a Laya/Jev-compatible API
layaFineTuning
codingpipeline
Fine-tunes Laya to route customer return requests and evaluates its accuracy and safety
the-new-turing-test
glowleaf
A Magic: The Gathering game harness tests Laya as a player
laya-banking77
hari-hashing
Evaluates Laya decision-model classification and routing on the Banking77 intent dataset
laya-bias-audit
maheshwari2000
An audit measures option-order, key, and wording biases in Laya decision-model checkpoints
HeXA-laya-demo
queso-gato1355
A local dashboard demonstrates Laya decisions for an autonomous delivery robot scenario
hybrid-decision-agent
musaugurlu
Tests a Laya-first agent architecture that routes requests to Gemini Flash and benchmarks its performance
Laya_Jev_opensource_guide
gotree94
Explains how to self-host Laya and Jev, including setup, fine-tuning, and benchmarks
laya-vs-jev-toxicity
jmgb27
Compares Jev and fine-tuned Laya on toxicity classification using Civil Comments
distill-public
abhi-sama
Fine-tunes local Laya decision models from synthetic data and evaluates and exports calibrated models
laya-demo
bbaotran
Snake game played by Laya on CPU, with Jev and other decision backends for comparison
jev-laya-lab
KarlPeixoto
Explores Laya and JEV decision-making through Python experiments and evaluations
laya-local-benchmark
saksham-personal
Benchmarks and packages offline CPU Laya decision workflows for Windows VDI environments
Sytem-1-model-testing--jev--Laya-
dinod001
Benchmarks System 1 decision models against autoregressive LLMs for latency, cost, and accuracy
system1-prompt-lab
msmidhunms
Evaluates and compares Laya and other System 1 classifiers with datasets and prompt optimization
LayaAI
Abii9443
Evaluates chatbot responses using Laya decision models and supports fine-tuning
laya-vs-jev
jupiutrera
Compares Laya and Jev confidence calibration on typed decisions using a pixel-art visualization
RightDesk-Laya
Supankan
Evaluates support-request routing workflows with a local Laya inference adapter and baseline comparisons
laya-korean-eval
sungreong
Measures Korean classification performance, latency, memory, and fine-tuning effects for Laya Multilingual
decision-model-benchmarks
asopitech-publishing
Benchmarks decision models including Laya and Jev on Japanese routing and support classification tasks
harmful-request-classification
zacharyabolton
Experiments compare TF-IDF, DeBERTa, and fine-tuned Laya for harmful-request classification
rag-benchmark
erendikmenn
A local Turkish RAG benchmark measures the contribution of Laya reranking
system-one-demo
dinh144
A research demo compares Laya, Kev, and small language models for agent-memory decisions
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