Why your local LLM feels dumber than it is

High-Performance Computing Machine Learning, LLMs, & AI
5

SIDE QUEST

5090 quant buyers guide

I took a small subset of results as the larger corpus coalesces and spit out a quick 5090 quant buyers guide: Qwen 3.8 Quant Selection Guide for RTX 5090

model routers: or, how to save your sessions for later testing and analysis

Also, for anyone wondering HOW you go about capturing real workflows for re-use in later testing, you just need a model router:

Work continues in the background. Stay tuned for the much larger readout on configs capabilities and costs.

16 Likes