5 results for “prompt caching”
OpenAI upgraded GPT-6 prompt caching with a 30-minute reuse window and new dashboard tools to monitor hit rates. The changes target persistent agent workloads and can cut cached input token costs by up to 90%.
Black Forest Labs released FLUX 3 Action, a 7B open-weight world action model hitting 38.3% on RoboLab-120 at up to 3.95× the speed of prior open models. Distilled checkpoints are included.
PyTorch Conference NA 2026 Core coverage expands with a Sept 16 Open Research, Tooling & Optimization guide — Nested Graph Breaks, Pyrefly, agent kernel pipelines, Monarch, NVFP4/HiFloat, and OpenEnv.
September 2026 watch through 22 Sep: GPT-6 Sol / Luna and Claude Opus 5.5 price drops, DeepSeek V4.1-Flash, Grok 4.7, MiMo-V2.6, Gemini 3.8 Live — Gemini 4 and Apple China timing still open.
OpenAI cut GPT-6 Sol and Luna API prices by 50% versus GPT-5.6 rates and published vendor benchmarks claiming they beat Claude Opus 5 and Fable 5.1 on cost-adjusted coding and agent tasks.