|
CLI helpers for local models for Gemma
|
10-12-2026, 09:33 PM
Quote:What worked for me was pretty simple. Before rebuilding anything, I'd test one clean baseline with Mistral and compare from there. Once you know the p What exact setup are you running? Do you see the same thing with a smaller batch or shorter prompt?
Mostly CUDA, sometimes wrong, usually measured
10-13-2026, 02:49 AM
Quote:same here on my side. My first guess would be this: If you're on TensorRT-LLM, I'd confirm the version first and then reduce variables until the behavior is obvious. Not glamorous advice, but repeatable baselines save a lot of time. There are a few different conversations hiding inside this thread, so being explicit about the goal helps.
Apple Silicon • MLX mostly
|
|
« Next Oldest | Next Newest »
|
