Typesafe AI launches Jev, a 20-200x faster frontier-level classifier trained with RLCD
soumitrashukla9 · x · 2026-09-17
Typesafe AI (after 2 years in stealth) released Jev, pitched as a new frontier AI model: an extremely fast and accurate classifier that takes predefined options plus a natural-language query/data and returns a solution with confidence scores.
- Claims 20-200x faster and 40-400x cheaper than frontier models (output tokens free), trained via a new method called RLCD
- Independent test by cramforce: against an existing eval that used Gemini 2.5 Flash Lite, Jev saturated the quality eval and ran 6x faster
- Community read: a "switch statement on steroids" capped at 255 options, useful for model routing, permission classifiers, safe/unsafe PR labeling, and as a cheap LLM-judge replacement
Caveat: vendor claims (including "co-invented ChatGPT") are unverified at scale.
More from coding & agent
- Monetize your MCP server: let agents pay via Stripe, already live at atomHQ and Donorbox — jeff_weinstein · 2026-09-17
- Supabase Select 26 lineup: YC's Garry Tan and Anthropic execs to speak in SF — garrytan · 2026-09-17
- Databricks rolled Astra out to all 3,500 engineers: beats Opus 5 on complex tasks, +60% coding spend — gdb · 2026-09-17
- Validating real-action agents is unsolved: one test run cost $30 and an X flag — Common_Dream9420 · 2026-09-17
- Texio: fail-closed Markdown section edits for coding agents, open-sourced under MIT — yjthegnius · 2026-09-17
- Agent Dev Pattern: Store a 'Consent Record' Before Every Agent Run — blaizedsouza · 2026-09-17