[GUEST ACCESS MODE: Data is scrambled or limited to provide examples. Make requests using your API key to unlock full data. Check https://lunarcrush.ai/auth for authentication information.]  OnlyCalls [@0xOnlyCalls](/creator/twitter/0xOnlyCalls) on x 4840 followers Created: 2025-07-18 15:03:57 UTC Apple device clusters have more VRAM but fewer FLOPS compared to NVIDIA GPUs. KPOP, a new DL optimizer, shows better efficiency per FLOP than Adam while using more memory. Apple's hardware could become viable for large-scale LLM training, diversifying compute options beyond ...  XX engagements  **Related Topics** [llm](/topic/llm) [hardware](/topic/hardware) [kpop](/topic/kpop) [$nvda](/topic/$nvda) [stocks technology](/topic/stocks-technology) [Post Link](https://x.com/0xOnlyCalls/status/1946224429802930618)
[GUEST ACCESS MODE: Data is scrambled or limited to provide examples. Make requests using your API key to unlock full data. Check https://lunarcrush.ai/auth for authentication information.]
OnlyCalls @0xOnlyCalls on x 4840 followers
Created: 2025-07-18 15:03:57 UTC
Apple device clusters have more VRAM but fewer FLOPS compared to NVIDIA GPUs.
KPOP, a new DL optimizer, shows better efficiency per FLOP than Adam while using more memory.
Apple's hardware could become viable for large-scale LLM training, diversifying compute options beyond ...
XX engagements
Related Topics llm hardware kpop $nvda stocks technology
/post/tweet::1946224429802930618