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
tetromino
tomayac
A browser tetromino game uses the Decisions API with selectable Laya and open-jev backends
system1-rerank-bench
dchristopoulos
A benchmark compares Laya and other decision models with rerankers on BEIR retrieval datasets
jev-websearch-eval
anweat
An offline web-search evidence evaluation compares hosted Jev with local Laya and a rule baseline
Reliability-of-Fine_Tuned-Typed-Decision-Models-in-Multiple_Choice-Tasks
PXShou329
Research code studies typed-decision fine-tuning, calibration, robustness, and transfer for Laya and other models
system-one-bench
alilibx
Benchmarks typed-decision models, including Laya, across labelled datasets for accuracy, calibration, latency and cost
laya-demo
foxnukem
Runs a Tower Bloxx browser agent that uses Laya to decide when to drop blocks
laya-governance-gate
yashkumar167272
Evaluates Laya decisions for schema-change governance and calibration
Laya-Dilemma-Lab
SherwinDesouza
Simulates cooperation decisions among 100 agents using the Laya model
laya-vs-jev
JinhoKim46
Compares Laya and Jev on decision quality, speed, and cost for support-ticket routing
laya-vs-reranker-benchmark
gowthxm07
Benchmarks Laya relevance filtering against cross-encoder reranking in RAG workflows
openai-decisions-vs-jev-vs-laya
mateusfonsek
Benchmarks OpenAI Decisions, Jev, and Laya on Portuguese-language decision tasks
mlx-quantization-lora-eval
D-Harshith
Benchmarks MLX quantization and LoRA alongside Laya encoder classifiers on GitHub issue classification
2048GameDecision
Devikrishna545
Compares Jev and Laya playing 2048 and fine-tunes Laya for improved accuracy
laya-autoresearch
Oqura-ai
A Python workspace for iteratively fine-tuning Laya typed-decision models against fixed evaluation contracts
laya_auto
diff-name-diffc
A live-web agent benchmark and fine-tuning workflow for Laya browser decisions
laya-mlx
Thar-Paing123
A native MLX runtime for running Laya typed-decision models on Apple silicon
laya-model-research
RoyalMilkteaMaster
Research code evaluates Laya as a control model for multi-hop question-answering loops
jev-laya-kev-winnow-qwen-comparison
srvkataria
Compares Qwen, Jev, Laya, and Kev on support-ticket routing
pi0-laya-experiments
must-I-jpg
Compares Laya action selection with baselines in reproducible π0 simulations on ALOHA and LIBERO-Long
laya-demo
vighnesh-chavan
Demonstrates Laya typed decisions in Doom and a custom pygame shooting game
FaradayLayaBenchmark
kurtfi
Benchmarks local models for MCP and agent security with a fail-fast evaluation pipeline
sys1-bench
ugly-custard
Benchmarks Laya and Von decision encoders and provides a practical Laya usage guide
btc-decision-map
Carolyn-ai
A BTC research dashboard compares local Laya decisions with a rule-enhanced Laya path and optional Jev
jev-vs-laya-vs-vega
janagarajsn
Benchmarks Jev, Laya, and Vega on synthetic customer-support ticket triage
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