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-steering-lab
Hantlowt
Tests methods for specializing frozen Laya models and benchmarks their accuracy and runtime
laya-training-log
ChenneyZhuang
Documents training and benchmark results for a fine-tuned Laya browser model
laya-first-look
cvranjith
Explores base Laya checkpoints locally on Apple M4 and reports informal behavior and latency measurements
Tenstorrent.Blackhole-convaiinnovations_laya
Thatch-cloud
Develops a Tenstorrent Blackhole inference backend and service integration for Laya
jev_and_laya_benchmarking
pavanjava
Benchmarks Jev and Laya accuracy and inference speed on typed-decision tasks
jev-laya-openai-comparison
amansahani
Benchmarks Laya and Jev against OpenAI models on financial regulatory decisions
decision-model-bench
SaiNarayana-B
Tests the accuracy and calibration of Laya decision-model confidence scores
jev-arena
eliot5566
Lets users create text-driven fighting bots piloted by Jev, Laya, or other models
laya-calibration-lab
BunsDev
Fits and evaluates probability calibration for Laya typed-decision models
laya-fp16-parity
xauberer93
Checks FP16 parity for the Laya model on a ZeroGPU lane
laya-trader
antonellof
Uses Laya MLX to paper-trade crypto and stocks with a dashboard and backtesting tools
system-one-reexamined
CodeWithMoin
Benchmarks Laya and Jev against zero-shot NLI and SetFit classifiers
jev-vs-laya
vizakan10
Compares Jev and Laya decision quality and speed by having both models play Wordle
LayaFinetuning
hermestwogetherbot
Tests Laya decision scenarios and documents a workflow for fine-tuning it on transcript events
laya-mlx-demo
jerryshao2012
Demonstrates an Apple silicon MLX conversion of Laya for typed decision inference
laya_finetuning
mkthoma
Fine-tunes Laya to normalise messy place records into a 10-class taxonomy, with phase reports
laya-tictactoe
amdisrar
Tests Laya's tic-tac-toe decisions against a Minimax reference
laya-snake
zhenda-hub
Runs Laya in Snake and compares its play with four other strategies
laya-classifier-finetuning
sallout
Fine-tunes Laya for custom text classifiers and optionally exports models to Core ML
jev-vs-laya-dino
RodolfoBonis
A Chrome Dino game demo compares Jev and Laya on gameplay, latency, cost, and performance
jev-laya-case-study
Takuakira
A controlled case study compares Jev and Laya decision responses as repair costs change
finetuning-laya-for-browser-agents
kp-algomaster
Code, datasets, and evaluations fine-tune Laya for structured browser automation decisions
frontdoor
debojitroy
Screens messages with Laya inference, specialization experiments, and reproducible evaluations
CatalogMesh
debojitroy
Maps supplier catalog products to marketplace categories using Laya inference
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