Compatibility check
Can I run Llama 3.1 8B Instruct on GeForce RTX 3080 10GB?
Yes — it runs.best quant:
Decode speed
125 tok/sest
Usable context
33K
Memory
10 GB
Fits in memory125 tok/sest · 33K ctx
Fits in memory107 tok/sest · 33K ctx
Spills to system RAM
