Providers quietly used cheap classifier routing for years while charging full LLM prices
sull · x · 2026-10-14
The author claims that before Jev's release, inference providers had already been leveraging classifier/decider routing behind the scenes — handling simple queries with cheap small models while still charging users for full LLM inference. He argues this should have been exposed as a proper service all along, meaning consumers were overpaying for years, and that @typesafeai's release forced the industry's hand.
More from Models
- Gemini 4 Argon Trails GPT 6 Astra by Just 25 Points on Physical Agents Index — qinzytech · 2026-10-15
- GPT-6-Astra scores 98% on unpublished theoretical physics problems, catching expert errors — sandersted · 2026-10-15
- Devs report quality degradation when Kimi K3 is served via third-party providers — casper_hansen_ · 2026-10-15
- Gartner's 2026 GenAI Hype Cycle: Agentic AI still climbing, AGI 10+ years out — luisdans · 2026-10-15
- Meta's Muse Wonder costs a quarter of Fable, but early testers call it slow with ugly UI — vista8 · 2026-10-15
- 5-person team open-sources frontier-class Ultramarine 1 Flash for ~$300k — willcb · 2026-10-15