OpenAI Cuts GPT-5.6 Sol API Prices by 50%; Bedrock Leads Latency
petrusenko_max · x · 2026-08-18
OpenAI has reduced GPT-5.6 Sol API prices by 50%, setting new rates at $2.50/1M input tokens and $15/1M output tokens. The model scores over 90% on coding and reasoning benchmarks with 99.29% availability. Amazon Bedrock currently leads in performance metrics for this model, recording 2.54s latency and 100 tokens/s throughput.
Related event: OpenAI Cuts GPT-5.6 Sol Price 50% on OpenRouter and Vercel AI Gateway(6 posts)→
More from Infra
- H100 hourly price hits $3.08 monthly high; 3-year-old chip still in demand — toptickcrypto · 2026-08-19
- Kimi K3 at Q1 writes working code in one shot; GLM5.3 API ran 7 hours and delivered — segmond · 2026-08-19
- The Real Cost of Agentic AI: Why Cache Misses Are Expensive — arena · 2026-08-19
- A $1 "hi": Why Agentic AI sessions get expensive when cache goes cold — arena · 2026-08-19
- AI agent platforms fail due to poor infrastructure, not prompts; OpenClaw blueprint reveals production architecture — aftahi_ai · 2026-08-19
- AI infra market analysis: Photonics surge 4.75%, Lumentum capacity sold out, Nvidia touts CPO efficiency — demian_ai · 2026-08-19