LLM API price spread across Claude, GPT, Gemini, GLM and Kimi hits ~3x at resellers
Life-Situation-1533 · reddit · 2026-10-29
The author priced the same three workloads (chat, RAG with 80% cache hits, agent loop with tools) across official APIs and 6 resellers on 2026-10-28. Surprises: Claude Sonnet reseller spread is 3x, not 10% (Gemini Flash 1.2x); cache pricing moves effective cost more than headline price (one provider discounts cache reads to 1/10, another doesn't list it); and the "cheap" resellers cluster on just 2-3 models, everything else at list price. Reseller prices change weekly and one of the six is the author's own.
More from Venture
- Betting on unsolved niche problems: a long-game playbook for AI startups — auto_grad_ · 2026-10-29
- FuriosaAI quadruples valuation to $2.9B in new funding round to challenge Nvidia — dinabass · 2026-10-29
- Thiel Fellowship's wild ROI: 40 under-20 fellows founded 13 unicorns, 4 are billionaires — FinanceYF5 · 2026-10-29
- Dev thanks 'shitty' GitHub Actions as his side project rakes in Stripe payments — aniketmaurya · 2026-10-29
- Modal Hits $1B Annualized Run Rate and Raises Series D Funding — charles_irl · 2026-10-29
- Old vs new users: AI natives skip Google, sites need both SEO and AI traffic — gefei55 · 2026-10-29