Benchmarks and comparisons with decision models

PostHog's 9B Jev-like model that reasons before deciding, trained with SFT and CISPO, with a diffusion drafter for speed.

Hypothesis-driven experiments characterising the Jev API, intended to inform a future football prediction product.

Comparative judgment of ASAP essays by Jev and two LLM judges, turning pairwise decisions into scores compared with human grades.

Learning experiment comparing Jev with an OpenAI model on GitHub issue ownership and on AG News, TREC and SNIPS classification.

Open Jev-style 0.6B decision model plugged into BlitzRank and Reranker-Guided Search, scoring k candidates per match in one pass.

Independent measurements of Jev versus open Laya on typed decisions, plus a local-first cascade that matches Jev at 1.8x speed.

Open MiniCPM5-2B decision model serving /v1/systemone: typed choice, noul and score probabilities in one forward pass.

Browser lab and TypeSafe-shaped API for asking typed questions of a Nemotron diffusion model and inspecting the distributions.

Open 150M encoder that answers noul, choice and score questions in one pass, trainable on a free Colab T4 in about 30 minutes.

Verdict: 151M ModernBERT/GLiClass non-autoregressive decision models with calibration heads, reporting wins over Jev and Laya.

Reproducible MLX and Colab recipes that turn Qwen decoders into Jev-like typed decision models, from 0.5B on a laptop to a noncausal 27B.

Chinese multimodal moderation research pairing a Qwen3-VL LoRA evidence extractor with Jev policy decisions, benchmarked against baselines.

Reproducible test of SubQ Snap and Jev on 3,840 questions about a fictional store's event log from 2K to 360K tokens of context.

Harness where Jev or Laya pick DOOM tactics and targets from legal options while a deterministic controller presses buttons.

Open Apache-2.0 browser-agent decision model (Qwen3.5-35B-A3B fine-tune) that doubled Jev 1.13's task success in the same agent.

MIT-licensed benchmark and leaderboard for Jev-class decision models: accuracy, calibration, latency and cost on sealed tiers.

Open benchmark for System One decision models on Choice, Score and Noul, reporting accuracy, calibration and robustness separately.

Convex's eval repo with a 108-question multiple-choice benchmark comparing Jev's Choice answers with 14 LLMs on accuracy, latency and cost.

Reproduction kit for the Decision Index leaderboard: runs 37 public benchmarks against any /v1/systemone engine.

System One Mosaic Benchmark comparing Jev and open decision models across 137 English and 27 Japanese Choice, Noul and Score tasks.

CUDA-native engine that serves open Qwen3.5 checkpoints behind a Jev-compatible API, reading option logits without decoding.

Library, CLI and app that classify PDFs, DOCX and PPTX with Jev and split document packets at predicted boundaries.

DeepSeek Harness plugin with jev_decision tools, an opt-in risk and loop auto-guard, and a live self-test benchmark against the Jev API.

Python tool that measures where to route decisions between Jev and a larger model on your own data, then routes by confidence.

Standard-library Python workbench for repeatable classifier latency and accuracy experiments across Jev, Laya and classifier.dev models.

Browser driver where Jev makes every decision and agent-browser performs every action, benchmarked against Claude Code on 10 tasks.

Jev-powered browser MCP server that makes each step in one parallel call and escalates to the driving LLM only when unsure.

Auto-clipper for talky videos: code enumerates every possible cut point and Jev judges where each standalone clip starts and ends.

grep by meaning: pipe text in, ask a yes/no question in plain English, and Jev returns matching lines, even behind tail -f.

Zero-shot named entity recognition that turns Jev's Choice answers into entity spans via nominate, verify and boundary questions.