Opus 5.5 and GPT-6 shipped 101 minutes apart, both cheaper
airesearch12 · x · 2026-09-27
The ThursdAI podcast (Sep 24) recapped a wild week:
- Timing: One week after lab heads signed a "pace the frontier" pledge, Anthropic and OpenAI shipped big models 101 minutes apart — Claude Opus 5.5 (Fable-level, 40% cheaper than Opus 5) and GPT-6 Sol & Luna (half of GPT-5.6's price).
- Pricing: Opus 5.5 is $4 input (was $5), cached reads cut to $0.20, $20 output per 1M tokens; GPT-6 Sol is half Opus 5.5's per-token price and Luna is 40x cheaper again. For agentic coding, cache reads dominate the bill — run your own workload.
- Elsewhere: Meta put Muse at the center of everything, Grok moved into Teslas, open source cloned Jev within a week, and the episode was produced by an AI.
A related note: JevBench went from 0 to 70+ models in a week, and someone even tried to hack the author's machine to steal the benchmark.
More from Models
- Fable 5.2 Leads MazeBench Token Efficiency; Billion-Token Runs Underway — patience_cave · 2026-11-16
- Fable 5.2 Tops MazeBench With 21% After 90 Hours and $6,000 in Credits — patience_cave · 2026-11-16
- Tencent Hy4 flash and InternLM Sage-S2.1-Flash both released, tool-calling tests requested — MajesticPuzzler · 2026-11-16
- User Shares Coral Reef Creation Labeled 'Qwen 4 Max (xhigh)', an Unannounced Model Name — intellectronica · 2026-11-16
- Stealth Model 'silicon-2' Surfaces in Devin, Suspected to Be Anthropic's Next Model — kimmonismus · 2026-11-16
- Google hides Gemini model label in AI Mode, rolls Gemini 4 Carbon to Pro subscribers — gaganghotra_ · 2026-11-16