Topic

RTX 5090 llama.cpp runs of Qwen3.8-27B Q4_K_M

All digests tagged RTX 5090 llama.cpp runs of Qwen3.8-27B Q4_K_M

RTX 5090 vs M5 Ultra: Which AI Powerhouse Wins?

Comparing the M5 Ultra and RTX 5090 for local LLMs. We benchmark Qwen3.8-27B to analyze TFLOPS, memory bandwidth, and token generation speed (MTP on/off).

Read article