jev-codex-router is live · free · last verified 6 Nov 2026 · viability monitor
jev-codex-router
Per-turn Codex model routing: Jev picks the model, thinking depth and speed mode for every call.
The routing_policy.py contract is the actual product: the four-question decision model and the mandatory-Astra triggers are cleaner than what most commercial LLM routers publish. Since it was archived read-only on Sep 24, 2026, treat it as a reference implementation to fork, not a dependency to install and forget. Expect to read code, not file tickets.
Verdict generatedVerified 8d ago · liveness 60/100 · cite: rightaichoice.com/tools/jev-codex-router
- Developers running Codex heavily who want routine edits and mechanical follow-through off the top tier
- Engineers comfortable running a local router on :4202 plus jev_server.py on :4319 behind LiteLLM
- Anyone building a Codex or LLM router who wants a legible, forkable policy contract to copy
- Teams that need routing applied to tool-call continuations, not just the first turn of a task
- Anyone expecting a hosted, zero-setup product with support
- Buyers who need a measured Codex quota-savings figure before adopting — the −60% figure is an old-policy simulation
- Teams that need an actively maintained upstream — archived read-only Sep 24, 2026
We scan live Reddit threads, YouTube comments, X posts, G2 reviews and other communities — and hand you an honest verdict in under a minute.
- Honest verdict, not marketing
- Real pros & cons from real users
- Attributed quotes with receipts
3 free scans · no card needed
Skip Jev Codex Router if you need a maintained upstream or a vendor to call — the repository was archived read-only on Sep 24, 2026, and there is no SLA, support contract or release-notes feed behind it.
Running it means operating a local stack — Codex Router on :4202, jev_server.py on 127.0.0.1:4319 and LiteLLM — so your compute and maintenance time are the real bill, not the MIT licence.
The whole project is MIT-licensed at $0, so it is cheaper than any commercial Codex routing or gateway product you would put in the same slot. The cost is not the licence but the operating overhead: you supply the local server on :4202, jev_server.py on :4319 and LiteLLM. It fits individual developers and small teams already running that stack, and it is a poor fit for organisations that need a supported, maintained dependency.
In short
jev-codex-router — Per-turn Codex model routing: Jev picks the model, thinking depth and speed mode for every call. Best for Developers running Codex heavily who want routine edits and mechanical follow-through off the top tier, Engineers comfortable running a local router on :4202 plus jev_server.py on :4319 behind LiteLLM, Anyone building a Codex or LLM router who wants a legible, forkable policy contract to copy. Free to use.
What is jev-codex-router?
jev-codex-router — Per-turn Codex model routing: Jev picks the model, thinking depth and speed mode for every call. Best for Developers running Codex heavily who want routine edits and mechanical follow-through off the top tier, Engineers comfortable running a local router on :4202 plus jev_server.py on :4319 behind LiteLLM, Anyone building a Codex or LLM router who wants a legible, forkable policy contract to copy. Free to use.
Jev Codex Router is an MIT-licensed, self-hosted monorepo that sits between Codex and the model backend and picks a model plus a thinking effort for each individual model call — including the continuations after tool calls, where session-level routers typically let one tier ride the whole task. It targets developers already pushing Codex hard who want routine edits and mechanical follow-through to stop drawing on the top tier without hand-editing settings each session. The request path is explicit rather than a black box: Codex points at Codex Router on :4202, the jev/auto route travels through LiteLLM to a local jev_server.py on 127.0.0.1:4319, which returns a model-and-effort pair from the Luna → Terra → Sol → Astra ladder. Jev answers four Choice questions in one request: whether the call falls under the mandatory Astra policy (architecture, independent final code review, risk-focused review), the cheapest sufficient capability tier, the minimum sufficient effort from low to max, and a bounded route lease of one_call, tool_chain or user_turn. Routing logic lives in server/routing_policy.py instead of behind a dashboard, so you can read the contract rather than trust it. It's Responses in, Responses out, with the SSE stream relayed verbatim so tool calls, reasoning and compaction behave natively. Operational guardrails ship in the repo: fail-open so any Jev error keeps the turn alive, a sentinel-file kill switch that routes without Jev instantly, quota fallback that fires only on observed native quota exhaustion, and a local decision log per routed turn at ~/.codex/codex-router/jev-router-live.jsonl. The README cites a historical simulation of roughly −60% versus full Astra on 237 turns under the old policy — and says plainly that this is not measured Codex quota saved and not evidence for the current policy. The repo was archived read-only on Sep 24, 2026 (274 stars, 51 forks, 69 commits). Compared with hosted LLM routers that sell a dashboard and support
Key features
- Per-call model selection for Codex, including continuations after tool calls
- Thinking effort from low through max chosen together with the model
- Four Choice questions per request: Astra policy, capability tier, effort depth, route lease
- Tier ladder Luna → Terra → Sol → Astra with routing priors per tier
- Mandatory Astra policy for architecture, independent final code review and risk review
- Route leases bounded to one_call, tool_chain or user_turn
- Standard speed enforced on every route, overriding an incoming Fast setting on retries
- Responses in, Responses out — SSE stream relayed verbatim, no format conversion
- Two projections: Jev sees bounded decision state, the model gets the canonical Codex replay
- Embedded router exempts jev/auto from conversation windowing and tool-result aging
- Fail-open — any Jev error keeps the turn alive on a safe fallback route
- Sentinel-file kill switch that routes without Jev instantly
- Quota fallback on observed native quota exhaustion, retried only with a distinct candidate
- Local decision log per routed turn for policy calibration
- Policy split-v11-native-first-fallback judges remaining work rather than inheriting a review's category
Behind the verdict
Where this repo earns its keep is the shape of the decision, not the savings number. Splitting model choice into four separable questions — Astra policy, capability tier, effort depth, route lease — means you can argue with one part of the policy without rebuilding the rest. Most routing tools we look at bury that in a weights file and a dashboard. Pick it when you're running Codex hard enough that routine edits and post-tool follow-through are visibly eating your top tier, and you're willing to run a local router on :4202 plus jev_server.py on 4319 behind LiteLLM. The guardrails make that palatable: fail-open keeps the turn alive, the sentinel-file kill switch bypasses Jev instantly, and the per-turn decision log gives you something to calibrate against instead of vibes. Pass if you need a supported dependency. The owner archived the repository on Sep 24, 2026, so it's read-only — no releases, no security patches, no one to escalate to. Any bug you hit is yours. That rules it out for organizations carrying a vendor SLA or compliance obligations. Also pass if you want a measured quota-savings figure before adopting. The −60% number on 237 turns is a historical simulation under the old policy, and the README says so plainly; it is not evidence for the current policy or a promise about your bill. Closest alternative: a hosted LLM router that gives you a UI and a support inbox in exchange for an opaque policy. If what you actually want is a legible contract you can fork and audit, this is a better artifact than most paid ones — just budget for owning it. Caveat in practice: the fixed local ports and the LiteLLM hop are non-negotiable infrastructure. Locked-down laptops and networks that can't open them will fight you the whole way in. We'd reach for this as a
Limitations
- The repository was archived by its owner on Sep 24, 2026 and is now read-only; no further updates, releases or maintenance are expected.
- The live scrape returned only generic GitHub site chrome (navigation, plans, features, whats-new) rather than project-specific docs, so detailed configuration and setup could not be re-verified.
- The evidence does not name any underlying AI model, so model-specific capabilities are unverified.
as of 2026-10-13
Researching jev-codex-router? Get your full AI stack in 60 seconds.
Free, no signup — tell us your goal and get tools matched to your budget & existing stack.
Is jev-codex-router still active?
jev-codex-router is live · free · last verified 6 Nov 2026 · viability monitor.
Viability score
How well maintained and how widely used is jev-codex-router? Built from what the vendor actually publishes (docs, changelog, tutorials, integrations, pricing), whether the site is live, and how much real users discuss it. How we calculate this
Last calculated: November 2026
How we score →What do users say about jev-codex-router?
How we verified this
Every listed fact about jev-codex-router — pricing, features, links, status — is re-read from the vendor's own pages by our verification pipeline and each change is logged with a date; the last pass ran on 6 Nov 2026. Analysis is drafted with AI models and checked against our editorial standards before publishing; see how we verify.
Verification history
We have re-verified jev-codex-router 6 times since . Each pass re-reads the vendor's own pages and re-checks every listed field against that evidence; passes where nothing had changed are marked as such.
- — changed pricing page
- — changed features, summary, pricing tiers, our verdict, description
- — changed pricing page, limitations
- — changed our verdict, summary, description, who should skip it, pricing tiers, features, who it suits
- — changed limitations, pricing page
- — changed features, our verdict, description, who it suits, pricing tiers, who should skip it
Free to cite with attribution — this page re-verifies continuously.
How much does jev-codex-router cost in 2026?
jev-codex-router: Free to use. Pricing last verified 6 Nov 2026.
Plans compared
Every published jev-codex-router tier, its price as the vendor publishes it, who it fits, and what it adds over the tier below. The 12-month cost projection follows.
| Plan | Price | Billing | Who it's for |
|---|---|---|---|
| Open Source (MIT) | Free | — | Developers already running Codex heavily with a local LiteLLM and jev_server.py setup, who want per-turn model and thinking-depth routing without a vendor contract.Adds: Starting tier and the only tier: $0 under MIT, covering the embedded router fork, fail-open routing, sentinel-file kill switch and the local decision log. |
12-month cost
Project the real annual outlay, including the implied monthly cost when only an annual tier is published.
Vendor list price only. Add-on usage, seat overages, and contract minimums are surfaced under Hidden costs & gotchas.
Where the pricing makes sense
The company stage and team size where jev-codex-router's pricing actually pencils out — and where peers do it cheaper.
The whole project is MIT-licensed at $0, so it is cheaper than any commercial Codex routing or gateway product you would put in the same slot. The cost is not the licence but the operating overhead: you supply the local server on :4202, jev_server.py on :4319 and LiteLLM. It fits individual developers and small teams already running that stack, and it is a poor fit for organisations that need a supported, maintained dependency.
Has jev-codex-router's pricing changed?
No plan price for jev-codex-router has moved since we began logging in Aug 2026; the pricing page is re-read on every verification pass.
Who should pick jev-codex-router?
Pick jev-codex-router if you are Developers running Codex heavily who want routine edits and mechanical follow-through off the top tier, Engineers comfortable running a local router on :4202 plus jev_server.py on :4319 behind LiteLLM, Anyone building a Codex or LLM router who wants a legible, forkable policy contract to copy, Teams that need routing applied to tool-call continuations, not just the first turn of a task.
Look elsewhere if you are Anyone expecting a hosted, zero-setup product with support, Buyers who need a measured Codex quota-savings figure before adopting — the −60% figure is an old-policy simulation, Teams that need an actively maintained upstream — archived read-only Sep 24, 2026, Organizations requiring a vendor SLA, compliance certifications or ongoing security patches.
Skip it if Skip Jev Codex Router if you need a maintained upstream or a vendor to call — the repository was archived read-only on Sep 24, 2026, and there is no SLA, support contract or release-notes feed behind it.
GeneratedReal-world workflow fit
Concrete scenarios for the personas jev-codex-router actually fits — and what changes day-one when you adopt it.
You install the monorepo, run Codex Router on :4202 and jev_server.py on 127.0.0.1:4319, then start a multi-hour refactor where most turns are trivial file edits and a few touch architecture.
Outcome: Each model call gets its own model and thinking-depth pick, with the Astra policy forcing the expensive tier for the architecture turns, and every decision is written to ~/.codex/codex-router/jev-router-live.jsonl so you can inspect why a route was chosen.
You run the documented BACKTEST.md protocol against your own work patterns, then check the decision log against what you would have picked by hand.
Outcome: You get a written protocol and its limitations rather than a vendor benchmark, so you can decide whether the policy split-v11-native-first-fallback behaviour matches your risk tolerance before anyone depends on it.
You hand AGENTS.md to a coding agent and let it perform the router installation end to end.
Outcome: The monorepo is self-contained — the maintained Codex Router fork is embedded under router/ with provenance in ROUTER_FORK.md — so no second Git checkout, submodule or hidden source clone is needed.
Use cases
- Route each Codex turn to the least expensive model that can still make the next decision.
- Reduce Codex quota consumption during long agentic coding sessions with many tool calls.
- Keep routing decisions active across tool-call continuations instead of only the first turn.
- Constrain the selectable model pool via the curated-model extension point.
- Hand AGENTS.md to an AI agent to perform the router installation end to end.
- Run the documented BACKTEST.md protocol to sanity-check routing behaviour before rollout.
Setup time & first value
How long it actually takes to get something useful out of jev-codex-router — broken out by persona, not the marketing-page minute.
Solo developer already running Codex and LiteLLM: roughly an hour to stand up Codex Router on :4202 and jev_server.py on 127.0.0.1:4319 and see the first routed turn in the decision log. Agent-assisted install driven by AGENTS.md: faster, but you still verify the two local endpoints. Team rollout: budget a backtest pass from BACKTEST.md before anyone depends on the policy, and remember the repo
Switching to or from jev-codex-router
How to bring data in from common predecessors and how to get it back out — written for the switcher, not the buyer.
- →From stock Codex with no router: point Codex at Codex Router on :4202 and let the jev/auto route pass through LiteLLM to jev_server.py on 127.0.0.1:4319.
- →From a hand-tuned per-session model setting: replace the manual choice with the four Choice questions answered per model call, including continuations after tools.
- →From a generic LLM gateway: keep LiteLLM and attach Jev through its generic-provider and curated-model extension points instead of rewriting your routing.
- ↗To stock Codex: drop the router from the path and use Codex's native models against the ChatGPT backend again.
- ↗To another routing layer: the embedded Codex Router fork under router/ is documented in ROUTER_FORK.md, so you can trace what was changed before moving on.
- ↗To a supported commercial gateway: expect to re-implement the Astra policy, the Luna → Terra → Sol → Astra ladder and the route-lease bounds yourself.
Integrations
Resources & guides
- Resourcegithub.com
AGENTS · jev-codex-router
Helpful link from github.com
- Resourcegithub.com
BACKTEST · jev-codex-router
Helpful link from github.com
- Resourcegithub.com
ROUTER FORK · jev-codex-router
Helpful link from github.com
- Resourcegithub.com
README · jev-codex-router
Helpful link from github.com
- Resourcegithub.com
CONTRIBUTING · jev-codex-router
Helpful link from github.com
- Resourcegithub.com
SECURITY · jev-codex-router
Helpful link from github.com
- Resourcegithub.com
CLAUDE · jev-codex-router
Helpful link from github.com
Tutorials & Learning
YouTube returned 6 videos for “jev-codex-router”, and we withheld 6: 6 did not mention jev-codex-router. We are showing none, because we could not prove any of them are about jev-codex-router.
Official links
What are the best jev-codex-router alternatives?
The closest alternatives to jev-codex-router we list are NadirClaw, Pioneer and dari.dev. Ranked by direct product-type match, not category overlap.
| Alternative | Pricing | Starts at | Last verified |
|---|---|---|---|
| NadirClaw | Freemium | Pricing not published | — |
| Pioneer | Paid | Pricing not published | — |
| dari.dev | Freemium | Pricing not published | — |
Alternatives to jev-codex-router
View all
Tools that pair well with jev-codex-router
Common stack mates teams adopt alongside jev-codex-router, with the specific reason each pairing earns its keep.
NadirClaw
LLM router that picks the cheapest Claude, GPT or Gemini model per prompt and per Claude Code or Codex turn.
Pioneer
Pioneer is a self-improving inference API that routes every call to the right model and retrains itself from your production traffic.
dari.dev
LLM routing that picks the cheapest model still hitting your quality bar, without stalling the agent loop.
jev-codex-router vs its closest rival
jev-codex-router's closest head-to-head is with Roo Code — These two tools solve entirely different problems and should not be shortlisted against each other. If you need a working, supported coding assistant today, neither is a safe bet: Roo Code is pre-launch with no pricing, demo, or downloadable extension, while jev-codex-router is a free, self-hosted Codex routing layer for engineers who already run Codex heavily and are willing to install and maintain GitHub-sourced tooling. Choose Roo Code only if you want to evaluate a future multi-agent VS Code assistant and accept unfinished software; choose jev-codex-router only if you run Codex often enough that per-turn model and effort routing can reduce quota burn and you can work from a README, install.sh, and markdown docs. Verdict generated
Head-to-head comparisons
Jev Codex Router vs Roo Code
These two tools solve entirely different problems and should not be shortlisted against each other. If you need a working, supported coding assistant today, neither is a safe bet: Roo Code is pre-launch with no pricing, demo, or downloadable extension, while jev-codex-router is a free, self-hosted Codex routing layer for engineers who already run Codex heavily and are willing to install and maintain GitHub-sourced tooling. Choose Roo Code only if you want to evaluate a future multi-agent VS Code assistant and accept unfinished software; choose jev-codex-router only if you run Codex often enough that per-turn model and effort routing can reduce quota burn and you can work from a README, install.sh, and markdown docs.
GeneratedJev Codex Router vs Bito
These overlap on only one axis — model routing for Codex — and diverge on everything a buyer cares about. Pick Jev if you are one developer burning Codex quota, you are comfortable working from a GitHub README and install.sh, and you want a free per-turn decision without a sales call. Pick Bito if you lead multiple teams running agents on multi-repo codebases, need one admin view of tokens, spend, and routing decisions, and require SOC 2 Type II, on-prem deployment, and no code storage. The catch on Bito: no published Governor or AI Architect usage rates and no same-day rollout, so budget a real scoping period. The catch on Jev: no SLA, no live measured savings number before adoption, and no manual per-turn control — every route uses standard speed.
GeneratedJev Codex Router vs Temporal Ai
These aren't alternatives — you'd never pick one instead of the other. Temporal is infrastructure you buy (or self-host) so AI agents and business processes survive crashes, retries, and sessions abandoned mid-run, with Signals, Updates, durable Timers, Schedules, and Saga compensation doing the heavy lifting. jev-codex-router is a free GitHub-sourced router that shaves Codex quota by making a model-plus-thinking-effort choice on every call, including continuations after tool calls. If your agents keep dying at step 40, that's Temporal. If your Codex bill is the problem and you're happy maintaining local tooling, that's the router — and you could run both, with Temporal keeping the agent alive and the router picking models inside it.
GeneratedJev Codex Router vs Dbos
These are not competitors, and you should not be choosing between them. Pick jev-codex-router if your pain is Codex quota burn and you want per-turn routing you can install from a GitHub README and kill with a sentinel file; be ready to live without SLA, support, or a measured savings number, and note it never lets you pick thinking depth manually. Pick DBOS if your pain is workflows dying mid-run and you already run Postgres — it solves crash recovery, queues, and cron there, with the open-source Transact package as the entry point. If you are budgeting for one, they do not compete for the same line item.
GeneratedFrequently asked questions
Categories
Best-of guides
For makers of jev-codex-router
- Get a Verified Listing for jev-codex-router ($39): a written decision within 1 working day.
- Free listing for another tool: Free reviews are decided in order, within 3 weeks of the review starting.
Used jev-codex-router? Help shape our editorial sentiment research.