Deel benchmarks Jev vs frontier LLMs: up to 59x cheaper, accuracy jumps
multiply_matrix · x · 2026-09-25
HR company Deel benchmarked TypeSafe's Jev against frontier LLMs across multiple use cases: up to 59x cheaper, up to 4x faster, with quality at parity or better on 6 of 8 checks. Accuracy actually surged — repeat questions went from 70% to 97%, expense categorization from 50% to 86%, and escalations caught the same cases with fewer false alarms. TypeSafe opened Jev to Deel just Sunday.
More from Venture
- Eclipse turned $147M into a $2.5B Cerebras stake, a 17x return, CEO Lior Susan reveals — Sethwinterroth · 2026-09-25
- Booking, Expedia, Airbnb down ~20% in a month as AI agents spark a 'Consupocalypse' — brucemacv · 2026-09-25
- Goldman: Big Tech AI capex to jump 50%+ to $1.2T in 2027, $1.4T in 2028 — rohanpaul_ai · 2026-09-25
- Microsoft's Enterprise OpenClaw Sparks Competition Fears: Meta Down 3.8%, Microsoft Up 3.4% — firstadopter · 2026-09-25
- Knowtex named prime awardee on VA's $775M ambient AI scribe contract — ycombinator · 2026-09-25
- Growth hacker builds tool to track who comes back after signup, not clicks — FellMentKE · 2026-09-25