via #Microsoft : Maia 200: The AI accelerator built for inference
https://ift.tt/CjP7Rem
#Maia200 #AIinference #AIAcelerator #MaiaSDK #Azure #Microsoft #Foundry #Copilot #OpenAI #GPT5 #LLM #SyntheticData #ReinforcementLearning #FP8 #FP4 #TSMC3nm #HBM3e #Datacenter #CloudCompu…

Maia 200: The AI accelerator built for inference - The Official Microsoft Blog
Today, we’re proud to introduce Maia 200, a breakthrough inference accelerator engineered to dramatically improve the economics of AI token generation. Maia 200 is an AI inference powerhouse: an accelerator built on TSMC’s 3nm process with native FP8/FP4 tensor cores, a redesigned memory system with 216GB HBM3e at 7 TB/s and 272MB of on-chip SRAM, plus...








