TthemsquaredBenchmarks & research

jev-benchmark (themsquared)

Reproducible benchmark for TypeSafe AI's Jev on agent tool-call risk classification: accuracy, latency, and whether the confidence score is worth routing on.

Explore repositoryVisit website

How it uses Jev

📖 Read the write-up: I Benchmarked Jev on Agent Tool-Call Risk. Calibration Held.

Source: community catalog. Details and metrics reflect the published community snapshot and are not a performance endorsement. Documentation details extracted from GitHub · the README · the project website.

Keep exploring

Similar to jev-benchmark (themsquared)

Browse all
jev-benchmarks project preview

jev-benchmarks

Benchmarks & research
Probability-aware evaluation for typed decision models: calibration, selective risk, latency, and reproducible benchmarks.
AMade byAbdelStark17
jev-phishing-bench project preview

jev-phishing-bench

Benchmarks & research
Jev (TypeSafe) vs Claude Haiku 4.5 on 2 000 phishing emails: accuracy, calibration, latency, cost. Reproducible benchmark.
AMade byanisselbd5
jev-rag-benchmark project preview

jev-rag-benchmark

Benchmarks & research
Reproducible benchmark for measuring Jev reranking quality, latency, and cost in RAG
EMade byerendikmenn14
jev-agent-skill-router project preview

jev-agent-skill-router

Agent tooling
Typed, confidence-aware agent skill routing with TypeSafe Jev.
GMade byGodsBoy17
typesafe-skill-router project preview

typesafe-skill-router

Agent tooling
TypeSafe (Jev) skill routing for Hermes Agent: names the one skill worth loading, before the model call. Opt-in, stdlib only, ~$0.001 per routed turn.
DMade byDECRUX981213
switchboard project preview

switchboard

Agent tooling
Guardrail + model router for LLM gateways on TypeSafe's Jev (System One model), with an independent accuracy/calibration/latency evaluation. Stdlib Python.
AMade byaniruddh-krovvidi0
jevper project preview

jevper

SDKs & clients
Jev-shaped (TypeSafe System One) classification wrapper over OpenAI-like clients: probabilities and confidence instead of prose
ZMade byzhulinchng12
typesafe-triage-guard project preview

typesafe-triage-guard

Applications
Three composable judgment pipelines on TypeSafe's Jev: support-ticket triage, observability alert triage, and a deploy-risk gate.
SMade byshivam2003-dev0
hermes-jev-skills project preview

hermes-jev-skills

Browser & computer use
Jev-powered model routing, memory, compaction, skill selection, computer and browser use for Hermes agents (also Claude Code and Codex)
KMade bykerpopule726