Efficient Request Queueing – Optimizing LLM Performance

NeoMME: an efficient Multimodal-native and Multilingual Encoder

BenchMIRT: What are LLM benchmarks actually measuring?

GigaPath-Flash and GigaTIME-Flash: Toward population-scale discovery with efficient pathology foundation models

What if pathology foundation models could do more with less? GigaPath-Flash and GigaTIME-Flash cut computational demands while maintaining strong performance, opening the door to larger studies and broader exploration. The post GigaPath-Flash and GigaTIME-Flash: Toward population-scale discovery with efficient pathology foundation models appeared first on Microsoft Research.

Granite 4.2 LLMs: How They're Built

A Blog post by IBM Granite on Hugging Face