What qualifies for the Benchmarks & research category?
Entries include datasets, evaluation harnesses, comparison studies, research prototypes, measurements, and reproducible experiments involving Jev.
bestjev homeAwesome Benchmarks & research Jev Projects
Explore an awesome collection of 99 verified open-source Benchmarks & research Jev projects on bestjev. Compare GitHub stars, languages, repositories, and implementations.
Last updated
All640Jev-like models18Today+110This week+155Official6SDKs & clients50Frameworks & integrations53Agent tooling158Browser & computer use51Applications72Games & simulations56Demos & playgrounds46
Benchmarks & research99
Other lists31ProjectLanguageAddedDescription
TianyuCodings/NanoJev
Benchmarks & research1,074PythonOpen 0.6B Jev replica with parallel decisions, complete probability distributions, training pipeline, weights, dataset, and live demos.
vinnylarouge/jevlike
Benchmarks & research1,008PythonTrain a small model that chooses among a changing list of text options, one probability per option in a single pass. Includes Doom, chess, and Wikispeedia demos.
hr98w/jev-visual
Benchmarks & research150PythonAn educational Jev-like visual inference experiment on Apple Silicon: shared context, direct candidate scoring, and local visual demos.
Mapika/decider
Benchmarks & research99PythonOne-pass typed decisions with calibrated probabilities (System One style model), fine-tuned from Qwen3.5-2B.
vinilana/jev-eval-agent
Benchmarks & research95HTMLPersonal-assistant agent built on Vercel's eve with 100 mocked tools, measuring how many steps it takes when Jev picks the tool versus the LLM.
kshetrajna12/reflex
Benchmarks & research79PythonA small open decision model: state + typed questions -> calibrated probabilities. A Jev / System One re-creation on Qwen3.5.
nekuda-ai/WindTunnel
Benchmarks & research69TypeScriptWebMCP benchmark comparing browser-agent interfaces, with Jev included as one of the evaluated configurations.
fstandhartinger/jevbench
Benchmarks & research67PythonJevBench v1 - a benchmark for Jev-class typed decision models: smart, cheap, fast, reliable, open.
daseinlabs/open-jev
Benchmarks & research56PythonOne-pass option scoring with a local Gemma 3 4B on Apple silicon via MLX, inspired by jevlike, with a Doom demo.
bnsd55/jevmlx
Benchmarks & research40PythonJev-style parallel constrained decisions for any MLX model on Apple Silicon. Typed, schema-valid JSON in one forward pass.
iammrduncan/typesafe-ai-benchmark
Benchmarks & research32TypeScriptThis is a LLM Gateway that mimics typesafe ai structured output. Like an imposter Jev.
Heman10x-NGU/Verdict-open-jev
Benchmarks & research28PythonNon-autoregressive decision engine on ModernBERT (151M) with calibrated uncertainty (RLCD), TypeSafe AI Jev benchmark audit, and in-browser WebGPU playground.
r-ms/mini-jev
Benchmarks & research24PythonMini-Jev: what a Jev-style typed-decision interface looks like on a frozen Qwen3-4B — read the option letter's logits instead of generating JSON. Preregistered experiment, results, teaching bench.
zhengxuyu/litjev
Benchmarks & research21PythonA reproduction of Jev that turns any Qwen model into a fast decision model, serving the same /v1/systemone schema (Choice, Score, Noul) with no training and no generated answer text.
rorshopping/jev-on-a-laptop
Benchmarks & research18PythonUnofficial study: Jev-style parallel typed decisions on stock 1.5B-8B models on an Apple Silicon laptop. Benchmarks, research notes, and a Hugging Face Space demo.
goodrahstar/jev-column-race
Benchmarks & research16JavaScriptJev vs Gemini 3.8 Flash: labelling 1,000 app reviews, 4.1× faster and 7× cheaper.
No matching projectsTry another keyword or category.
Category guide
What does the Benchmarks & research category include?
Benchmarks and research evaluate Jev behavior, decision quality, calibration, performance, and practical tradeoffs.
Entries include datasets, evaluation harnesses, comparison studies, research prototypes, measurements, and reproducible experiments involving Jev.
Benchmarks & research FAQ
How should I compare Benchmarks & research projects?
Start with the documented use case, repository activity, primary language, license, and visible Jev implementation. GitHub stars are a dated discovery signal, not a quality score.
How does bestjev verify Benchmarks & research entries?
bestjev checks public repositories, project-authored documentation, code, demos, and other visible evidence. Always confirm current behavior and compatibility in the upstream project.
Which sources support this category?
bestjev grounds this category in named primary sources: Official Jev documentation · Public GitHub repositories.
“Typed outputs that software can act on.”— TypeSafe AI