Comment on Are you friends with any AI bots?
brucethemoose@lemmy.world 1 day agoI’m in a 24GB 3090 + 128GB RAM.
With full 300B GLM 4.6, I typically run 12K-28K context with different settings. I could do more than 28K, but the higher quantization starts to become a problem. And I get 5-6 tokens/s text-generation doing that.
With GLM Air? I can get a lot more, closer to 64K.
With smaller models that’s no issue.
I only get 3-5 questions in before I run out of tokens.
IDK how you’re prompting it, but you should clear the thinking block after every question, and that should leave plenty of tokens.
What are your inference server settings?
quediuspayu@lemmy.dbzer0.com 9 hours ago
I’ve been curious about running LLMs locally. Mostly about how long it takes to return a response.
brucethemoose@lemmy.world 9 hours ago
I mean, it depends on your hardware and the model’s size/intelligence.
Worst case for me is many seconds of preprocessing followed by 4-5 words a second.
But you can get almost instant responses + way faster than you can read too.