Exactly, I’m in the same situation now and the 8GB in those cheaper cards don’t even let you run a 13B model. I’m trying to research if I can run a 13B one on a 3060 with 12 GB.
I also have a 3060, can you detail which framework (sglang, ollama, etc) you are using and how you got that speed? i’m having trouble reaching that level of performance. Thx
Thanks for the additional information, that helped me to decide to get the 3060 12G instead of the 4060 8G. They have almost the same price but from what I gather when it comes to my use cases the 3060 12G seems to fit better even though it is a generation older. The memory bus is wider and it has more VRAM. Both video editing and the smaller LLMs should be working well enough.
Exactly, I’m in the same situation now and the 8GB in those cheaper cards don’t even let you run a 13B model. I’m trying to research if I can run a 13B one on a 3060 with 12 GB.
You can. I’m running a 14B deepseek model on mine. It achieves 28 t/s.
Oh nice, that’s faster than I imagined.
You need a pretty large context window to fit all the reasoning, ollama forces 2048 by default and more uses more memory
I also have a 3060, can you detail which framework (sglang, ollama, etc) you are using and how you got that speed? i’m having trouble reaching that level of performance. Thx
Ollama, latest version. I have it setup with Open-WebUI (though that shouldn’t matter). The 14B is around 9GB, which easily fits in the 12GB.
I’m repeating the 28 t/s from memory, but even if I’m wrong it’s easily above 20.
Specifically, I’m running this model: https://ollama.com/library/deepseek-r1:14b-qwen-distill-q4_K_M
Edit: I confirmed I do get 27.9 t/s, using default ollama settings.
Thanks for the additional information, that helped me to decide to get the 3060 12G instead of the 4060 8G. They have almost the same price but from what I gather when it comes to my use cases the 3060 12G seems to fit better even though it is a generation older. The memory bus is wider and it has more VRAM. Both video editing and the smaller LLMs should be working well enough.
Ty. I’ll try ollama with the Q-4-M quantization. I wouldn’t expect to see a difference between ollama and SGlang.
I’m running deepseek-r1:14b on a 12GB rx6700. It just about fits in memory and is pretty fast.