Team builds Jev-powered MCP router that cuts model costs up to 90% with 76% speedup
Fun-Meaning-6474 · reddit · 2026-10-28
The ai/ml api team built an MCP router using Jev, a new fast and cheap classifier model that takes JSON in and outputs probabilities. Jev inspects each request and routes it to the right model — no paying Opus prices for short messages — and also picks among 1,000+ agent tools (scraping, YouTube transcripts, SEO data) through the same API.
Their tests showed up to 90% lower costs and a 76% speedup. Agents supporting remote MCP can plug in to access models from 70+ providers; the project is still collecting feedback and a name.
More from coding & agent
- Chasing the Karpathy Peak: researcher maps AI productivity's slop cliff — anshulkundaje · 2026-10-28
- USearch v3 RC ships after 1,000 days, its libraries now embedded in 550K-star projects — viglovikov · 2026-10-28
- In the agentic world, the lines between engineering disciplines blur — yunta_tsai · 2026-10-28
- LLMs supply intelligence, agents do the work — do SaaS interfaces become obsolete? — ThomasBuildLab · 2026-10-28
- Codex scans your connectors/MCPs unprompted in the background, and you can't tell — kenziyuliu · 2026-10-28
- UCLA and Tsinghua paper probes how far coding agents can go in visual generation — ucla-cmllab · 2026-10-28