Pin the Jev model id before you freeze the baseline — JEV Brief · Sep 24
Nothing TypeSafe-official moved the contract today. The costly quiet-day miss is still the same: you freeze a baseline on jev-latest, then invent a story when next week’s numbers move. If you only have four minutes, write down the exact model id you will pin, refuse floating aliases on anything that auto-acts, then scan Top Repos.
- Name the model before the scoreboard. An issue on jev-intent-review puts it cleanly: record a concrete Jev model identity before you lock benchmark baselines. If the run log only says “Jev,” you do not have a baseline — you have a mood.
- Pin the router, not the vibe. Prefer a dated id such as
jev-1.13.0overjev-latestanywhere Choice can auto-act. A floating alias on an eval harness is the same drift as a floating alias on a support triage bot. - Schema-safe ≠ comparable. Constrained Choice still picks wrong when the underlying weights moved. Keep irreversible tool calls and arithmetic in code, with an abstain / low-confidence branch before the executor runs.
- Fleet doors need a closed menu too. If you expose Jev over MCP or a CLI, ship named tools (
judge,rerank,systemone) over a free-form “ask Jev anything” prompt. The menu is yours; Jev ranks inside it.
Top Repos today
Live board: https://jev.pro/top-repos. Stars are public rank, not a quality award. Metrics still sit on the 2026-09-21 snapshot (midnight refresh has not moved), so 1-day deltas are flat and the tables show stars only.
Jev-Based
| # | Repo | Stars |
|---|---|---|
| 1 | browser-use/jev-ultrafast | 14,008 |
| 2 | tamaratran/fast-jev-compaction | 5,763 |
| 3 | jarrodwatts/jev-trader | 1,690 |
| 4 | typesafe-ai/skills | 1,382 |
| 5 | Sac-Y/Jev-cu | 520 |
| 6 | thruwire/foreman | 450 |
| 7 | devagrawal09/jev-review | 437 |
| 8 | superagents-lab/jev-search | 354 |
| 9 | dabit3/jev-experiments | 352 |
| 10 | kerpopule/hermes-jev-skills | 340 |
Jev Integrations
| # | Repo | Stars |
|---|---|---|
| 1 | OpenByteInc/QuantDinger | 11,879 |
| 2 | 0xPlaygrounds/rig | 8,691 |
| 3 | ax-llm/ax | 2,935 |
| 4 | Arize-ai/openinference | 1,226 |
| 5 | reticlehq/reticle | 783 |
| 6 | awlevin/typesafe-computer-use | 693 |
| 7 | AgentiLoop/Agent | 618 |
| 8 | agentjido/req_llm | 580 |
| 9 | fhshaik/typesafe-mario | 314 |
| 10 | rokbenko/quackd | 222 |
OSS alternatives — not TypeSafe Jev
| # | Repo | Stars |
|---|---|---|
| 1 | mizorewww/laya-mlx | 2,255 |
| 2 | TianyuCodings/NanoJev | 1,688 |
| 3 | jaredpalmer/kev | 1,483 |
| 4 | bespokelabsai/nimble | 1,385 |
| 5 | vinnylarouge/jevlike | 1,141 |
| 6 | featherless-ai/simple-jev | 430 |
| 7 | wfzyx/von | 296 |
| 8 | ekzhang/openjev-sglang | 244 |
| 9 | razorback16/openjev | 242 |
| 10 | Heman10x-NGU/openJev-verdict-2.0 | 221 |
Ultrafast still leads Jev-Based; QuantDinger still leads Integrations. Treat the alternatives track as local-door reading (Kev / NanoJev / von) — star rank is not a swap.
What moved
Nothing TypeSafe-official that changes the contract, model pin, price, or channel door.
Worth one click
- Obrais-cloud/typesafe-mcp — MCP server that exposes TypeSafe (judge / rerank / systemone) to a fleet. Use it as a door check: named tools, or a free-form “ask Jev” prompt?
- Lasimeri/Mechanical-Jev — Rust client +
mjevCLI + eval harness for System One (noul / choice / score). Client-side only; still pair it with an explicit model pin before you trust the harness numbers.
Read it here
Read on the blog. Also on jev.pro: the live board at https://jev.pro/top-repos. The 09:00 PT email is this same article in this same order.