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-deploy
iintothewind
Deploys and benchmarks Laya Serve on Windows with Docker and a temporary public tunnel
laya-lab
Parswanadh
Coordinates long-context decision-engine research for the open Laya model
laya-indic-bench
jay123anta
Evaluates Laya decision models on Hindi and Assamese
jev-laya-tetris
HarryReidx
Benchmarks TypeSafe Jev against local Laya in a competitive Tetris duel
jev-n-laya
jcezardasilva
Compares Jev, Laya, Von, and Gemma on typed-decision tasks
laya-lk-bench
mithilyr
Stress-tests Laya decision-model performance on Sinhala and Tamil offensive-language data
laya-vs-jev
alilibx
Benchmarks open-weight Laya models against TypeSafe Jev on labelled datasets
laya-cuda-bench
bhushankinge
Benchmarks Laya inference throughput, latency and cost across NVIDIA GPUs and serving backends
laya-cn-study
Adkid-Zephyr
Collects Chinese-first post-training experiments, datasets, and evaluations for Laya
laya-vs-jev-traffic
ameeetgaikwad
Compares local Laya and cloud Jev in a real-time traffic-control simulation
jev-laya-classification-bench
bhushankinge
Benchmarks Jev, Laya, and Qwen on classification of federal IT solicitations
von-laya-jev-paint-compare
zhangyunting123
Creates side-by-side paintings from typed decisions by VON, Laya, and Jev
laya-coreml-vs-jev-benchmark
sallout
Compares Laya and Jev on zero-shot intent classification benchmarks
laya-v2-agent-routing
mdad-elec
Fine-tunes a Laya System-One router to select the price tier for each agent turn
korean-decision-benchmark
jkf87
Benchmarks SemIf, Decider, Laya, and Jev on Korean hate-speech classification
synthetic-ehr-decision-evals
MadCodeTX
Benchmarks decision models on synthetic EHR administrative-routing tasks
rlcd_x_1b-4b_instruct
mrgonzales-dev
Tests RLCD models including Laya and explores their use with 2B–4B models
gyra
Gowtham-R-2002
Fine-tunes Laya for fast coding-agent decisions and evaluates it against labeled tests
jev-vs-llms
ivanviragine
Compares Jev, Laya, other open models, and chat LLMs on loaded questions in two languages
wizden-moonlander
WizdenOrg
A lunar lander game compares Laya and Jev as typed-decision flight controllers
small-decision-model-cross-domain-degradation
SnackTerminator
A controlled study evaluates cross-domain and cross-lingual transfer in Jev-class decision models
ai-experiments
nadeem4
A collection of documented AI experiments reports protocols and results for model evaluations
laya-playground
shwetankg07
Three browser games measure Laya’s decisions against exact ground truth
laya-scripts
muck-stump
Shell scripts test authentication, server behavior, concurrency, and routing for Laya-compatible APIs
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