jev-agent-failure-benchmark

Benchmarks and a playground for TypeSafe's Jev (System One) model: chess, and who-is-the-player-talking-to for speech-to-text game NPCs

Evaluating TypeSafe's Jev as a fast monitor and action gate for agent sabotage in SHADE-Arena, compared with Gemini 2.5 Flash/Pro.

Benchmark TypeSafe JEV against LLMs, fine-tuned BERT, Laya and zero-shot NLI on text classification: accuracy, calibration, latency, throughput, cost

A browser agent that turns a single goal into a sequence of lightning-fast actions.

Open alternative to Jev: typed, calibrated decisions from any open-weights LLM in one forward pass (HF + vLLM), with benchmarks

Official drop-in TypeSafeClient replacement backed by OpenAI, Anthropic, and compatible LLM APIs, for comparing Jev against chat models.

⚡ Browser agent that acts on a hunch: Jev (TypeSafe System One) picks every click in ~150 ms, an LLM is only needed when confidence drops. Zero-dependency Python on top of agent-browser.

When Jev meets LLM — Transparent proxy that makes AI coding agents 3× faster and 80% cheaper. Cut response times from 2.4s to 890ms with Line J architecture.

Replicating Jev with a local LLM