Model routing
Pick which downstream model or tier should handle a request.
49 rows are filed under this pattern, 48 of them with code. Rows are listed as the catalogue orders them: official first, then rows with code, then by stars.
Filter and search them in the catalogue →
Rows marked “not Jev itself” or “shadow mode” stay listed, but they are not examples of deciding with Jev.
-
Cookbook: Structured data extraction cascade
A two-stage mini-then-verify-then-reasoning cascade that reaches most of a big reasoning model's quality at a fraction of the cost.
-
Pattern: Intent routing
Classify an incoming request and route it to the cheapest adequate handler: deterministic code, a specialist LLM, or a person.
-
claude-code-templates: three Jev plugins
Three independently installable Claude Code plugins — guardrails, model router and skill suggestion — each with its own hooks and tests.
-
@langchain/typesafe
The JavaScript counterpart of the LangChain integration, with the same classifier and middleware shapes.
-
hermes-jev-skills
Nine agent skills plus a CLI covering model routing, memory filtering, turn retention, one-of-many skill selection and next-action choice.
-
jev-review
Pre-screens code review with Jev to surface high-risk changes for a more expensive model or a person, with a local dashboard.
-
jevrouter
A router for models, tools and subagents.
-
Astra-Ares
Runs a patched Codex CLI in which Jev reads bounded task context and, with two choices, picks the reasoning effort for GPT-6's next response and how many responses (1, 2, 5 or 10) it should last; GPT-6 can change effort without breaking the prompt cache. An experimental reference implementation.written by a model
-
jev-codex-router
Judges how hard a coding turn is, then picks the model tier, reasoning depth and speed mode to match.
-
jev-eval-agent
An agent that routes evaluation work through typed decisions.
-
jev-kit
Claude Code kit: a PreToolUse guard where a free pre-filter passes about 93% of tool calls and Jev judges the rest (allow, warn, rewrite or block; fails open), sub-agent sizing, and Jev wired into file search, browsing and review. The author reports 9/9 browser runs at 0.0008 USD of Claude spend.written by a model
-
stuntd
Local proxy that learns your app's typed LLM decisions and answers them with a Laya head. Jev and OpenAI compatible.upstream description
-
Jevstiller
Proxy in front of a repeated Jev classification call: it forwards to Jev at first, trains a small local model on Jev's own answers and probabilities, and answers locally once it agrees with Jev on held-out traffic. The author reports 61.7% of Banking77 answered locally at 99.4% agreement.written by a model
-
jev-use
An agent plugin that hands steps needing no text output to Jev instead of the main model.
-
jevonian
A local endpoint between a coding agent and its providers that routes each turn to the most cost-effective capable model, manages quota and cache, and logs every routing decision to a local ledger.
-
a3m-router
⚡ Adaptive multi-model LLM router — 80+ providers, Jev System One single-pass routing (model=jev-auto), pheromone-trail failover, parallel ensemble merge. npm: adaptive-memory-multi-model-routerupstream description
-
pi-jev-router
Automatic model routing for Pi using TypeSafe's Jev through Vercel AI Gatewayupstream description
-
jev-opus
Runs Claude Code on Opus 5.5 with Jev re-picking the effort level when a prompt arrives and after every tool batch (up after a failing test, back down once it passes), sent as per-message effort statements so the prompt cache survives. Works as a local gateway or a headless driver.written by a model
-
jev-router
Open-source LLM router that uses TypeSafe's Jev to pick a model, on top of LiteLLMupstream description
-
pi-jev-router
A minimal Pareto-optimal OpenRouter model router for pi, based on Jevupstream description
-
jev-engineering
Claude Code PreToolUse gate: regex hard rules and a read-only allowlist first, then one Noul and one Choice decide allow, ask or deny, so clearly safe commands skip the permission prompt. Codex, Cursor and OpenCode adapters, plus a review-router GitHub Action.
-
jev-router
LLM router that picks the cheapest model capable of handling a query, using TypeSafe's Jev for fast classification instead of an LLM call.upstream description
-
pi-jev
Pi extension suite powered by Jev: selective context compaction and model routingupstream description
-
slo-router
SLO-aware LLM inference router with Jev decisions, live queue metrics, counterfactual evaluation, and reproducible latency/cost benchmarksupstream description
-
Codex Jev Router
Uses Jev Choice and Noul judgments on short task summaries to select a Codex subagent model and reasoning effort with a Sol fallback.
-
jev-auto-router
Jev Auto Router (Jev Router): experimental per-call GPT model routing for Codex via TypeSafe Jev and a local Responses proxy, with independent task verification.earlier upstream description
-
jev-codex-bridge
Model and reasoning routing for Codex Desktop and CLI, with a Windows service and validated updatesearlier upstream description
-
dsh-plugin-jev-effort-selector
A DeepSeek Harness plugin that picks the reasoning level automatically: Jev judges how much thought each message deserves, follow-ups inherit the topic's depth, low confidence rounds up, and any failure falls back silently.
-
jev-model-router
Model router for Claude Code using Jevupstream description
-
jev-codex-pilot
Smart Codex overlay with JEV model routing, context optimization & Kanban automation. Reduce tokens, keep control
-
jev-smart-router
JEV Smart Router — a Databricks App that uses TypeSafe JEV to pick which model answers each message, then runs inference on the chosen Databricks Foundation Model API endpoint.upstream description
-
jev-gate
Not every coding task needs your best model. Experimental Jev-powered model routing for Claude Code — V3 prototype runs today, V4 routes at the task boundary.upstream description
-
jev-harness-router
Per-turn router for agent harnesses: one 350ms Jev call picks the model tier, effort, tools and skill, behind a hard deadline with a regex fallback. Claude Agent SDK adapter included.upstream description
-
smart-switch
Reimagined window switcher for macOS using frontier artificial intelligence. Predicted by TypeSafe's Jev modelupstream description
-
tiershift
Model router that starts each request on the cheapest tier and lifts it only when Jev's signals (multi-step reasoning, difficulty, stakes, safety) cross a YAML policy. On its author's 120 prompts it matched the flagship's quality at 40% less cost; the fast model alone did as well at 3% of the cost.written by a model
-
janus
Measure when to use Jev and other models on your data, then route accordingly.upstream description
-
jev-model-router
Jev (TypeSafe) model router on the Vercel AI Gatewayupstream description
-
jev-model-router
Pick the best AI model and reasoning effort for any task in ~1s. Plugin for Claude Code, Claude Desktop and Codex, powered by TypeSafe's Jev decision model and live OpenRouter pricing. Balance intelligence, speed and cost, or choose your priority.upstream description
-
jev-synthetic-survey
Jev vs GPT-4.1 as synthetic survey respondents on Twin-2K-500. How you ask mattered more than which model you used.upstream description
-
XavierJev
A local decision layer in Jev's shape: yes/no, choice and rubric questions read off one token's logprobs from a local model, with a Claude Code permission gate measured on held-out command sets.
-
hermes-jev-router
Experimental Hermes plugin: Jev-assisted model routing plans with budget and capability constraints. API access pending.upstream description
-
Jev Switch
Lightweight gateway that serves POST /v1/systemone and routes each call through an editable graph of upstreams (TypeSafe, Vercel, OpenRouter, Laya or a local server) with priorities and failover. Rust daemon with a React console, Tauri desktop app and Docker image. Chinese-first docs.written by a model
-
stuntdouble
A drop-in /v1/systemone proxy: the app keeps calling Jev while local decision models answer the same requests in the shadow, then it reports whether they would have decided the same — per question, per confidence band, at the app's own decide() — plus Brier, ECE, latency and cost.
-
switchboard
Guardrail + model router for LLM gateways on TypeSafe's Jev (System One model), with an independent accuracy/calibration/latency evaluation. Stdlib Python.upstream description
-
Building a Harness with Jev
LangChain's explainer and integration walkthrough: the three question types, plus model routing and gating risky tool calls before they run.
-
Jev AI Use Cases
Walks through use case after use case — agent routing, an in-agent decision layer, ticket triage — each with a concrete option set and a sample response.
-
langchain-typesafe
The LangChain integration: a classifier plus experimental middleware for model routing and for gating risky tool calls before they run.
-
pi-typesafe-router
A Pi extension that has Jev classify each request and route it to the right model, off until you configure the mappings.
-
Jev router: how model routing works, and 16 routers compared
Explains a router that asks Jev which model, tool or skill the next turn needs before anything is spent, why routing down and back up can cost more than not routing, and compares 16 routers for Claude Code, Codex and OpenRouter.