Skepticism regarding Mythos inference optimization claims
scaling01 · x · 2026-08-15
The author doubts the rumor that OpenAI optimized inference for the Mythos preview, making it 2.5x cheaper to serve. The argument is that this implies the model was trained inefficiently, wasting billions of dollars, and then served internally for two months in that unoptimized state, which seems unlikely.
More from Infra
- a16z Analysis: OpenAI GDM Hires Hit Zero in 2026 as Neoclouds Rise — morgymcg · 2026-08-15
- RTX 5090 Exceeds €5,000 in Europe as AI Demand Drives Consumer Hardware Prices — XFreeze · 2026-08-15
- India plans up to 8 new chip plants and 5 nuclear reactors targeting 100 GW by 2047 — Polymarket · 2026-08-15
- ollama-uplink: Access your local Ollama from anywhere — deepunderscore · 2026-08-15
- Plivo cuts India AI agent voice call price to ₹0.19 per 30 seconds — bevenky · 2026-08-15
- OpenAI's GPT-5.6 Sol on Cerebras is nearly 7x faster than Fable 5 — haider1 · 2026-08-15