@steeve: another 5 days later, zml/llmd runs fully on Metal, serving 8 simultaneous requests at full bf16 zml/llmd is our LLM se…
Summary
zml/llmd now runs fully on Apple's Metal API, serving 8 simultaneous requests at full bf16 precision, with continuous batching and other modern features.
View Cached Full Text
Cached at: 06/13/26, 02:17 PM
another 5 days later, zml/llmd runs fully on Metal, serving 8 simultaneous requests at full bf16
zml/llmd is our LLM serving software that all the modern niceties (continuous batching etc..)
ps: “full bf16”, what a time to be alive https://t.co/UtSae5UvNj
Similar Articles
this looks promising: stepfun-ai/Step-5-Preview-BF16 · Hugging Face
StepFun AI has released a preview version of their Step-5 model in BF16 format on Hugging Face, indicating a promising development in AI model accessibility and performance.
@RayFernando1337: OpenAI plz acquire Bitrig and add this to your app ASAP!
A tweet requests OpenAI to acquire Bitrig and includes a demo of an iPhone emulator built using Astra and Metal.
@rohanpaul_ai: Politico: Judge Mark Pittman blocked OpenAI from seeing Apple's confidential deal with X and SpaceXAI. For context, aft…
Judge Mark Pittman blocked OpenAI from accessing Apple's confidential deal with X and SpaceXAI, ruling it irrelevant to OpenAI's defense in a lawsuit by Elon Musk's companies. The settlement remains secret.
Apple M6 Pro Achieves the Highest Single-Core CPU Score in Geekbench 7
Apple's M6 Pro chip has achieved the highest single-core CPU score in the Geekbench 7 benchmark, demonstrating its leading performance capabilities.
@GergelyOrosz: Can we talk about how Apple and Google have been pacing the frontier for native iOS and Android apps with the App Store…
The tweet discusses how Apple and Google have maintained unchanged App Store review processes since 2008, impacting native iOS and Android app development.