x.com
amVT on X: @airesearch12 @sesigl @benchmarkheaven This is pretty good..
@cruzex100@cruzex100listed 50m agoreviewed by Jev@airesearch12 @sesigl @benchmarkheaven This is pretty good.. My smoke test earlier: Accuracy: Jev 0.727 Soft accuracy: Jev 0.580 · Laya 0.471 Calibration (ECE, lower is better): Jev 0.144 · Laya 0.213 Latency: Jev ~710ms · Laya ~30-40ms Cost per decision: Jev ~$0.0004 (API) · Laya ~$0 (self-hosted) ❤️ 1 likes on X
- Author
- @cruzex100
- Use case
- Open Source
- Added
- 2026-09-25
- Cost
- $0.0004
- Time
- ~710ms
All figures come from the author. Check the source before you quote them.
More in Open Source
- apolinario (poli) on X: @flaviusburca congrats on the release, i've added it to the jev decision index 0▲ 0x.com
- apolinario (poli) on X: @XyneAI congrats on the release! i've included xor on the Decision Index 0.2 ben▲ 0x.com
- GitHub - draeder/Vej: A local Jev: calibrated yes/no, choice and score judgments from small NLI models, in the browser or as a Node server.▲ 0github.com
- Ed Plese on X: Jev but for images? I built a local model to try it out! 63ms for a 512x384 imag▲ 0x.com63ms
- WquGuru on X: Jev 发布才 8 天，开源替身已经一堆了，今天诺基亚也来蹭热闹了😅▲ 0x.com25ms
- GitHub - backant-io/jevelry: Use Jev everywhere to make & track decisions▲ 0github.com
Jev guides for this use case
- Open-source Jev: what is actually open, and what is notThe model is hosted, the tooling is open — the exact split, plus every open-source Jev repo in the directory.
- Wiring Jev into Claude Code and Codex via MCPPutting a decision model behind a coding agent: what MCP buys you, what to gate, and what we have not verified.
- Is Jev reliable? The honest limitationsWhat the no-hallucination claim really covers, where the accuracy ceiling sits, and the failure mode we hit ourselves.
Bid history
No bids yet — the first one takes this project straight to the spotlight.