Re:DeepSeek-R1 with Dynamic 1.58-bit Quantization
by danielhanchen on Tuesday January 28, 2025 @05:48AM
(#42850938)
The good thing is since MoEs are mainly memory bound, we just need (VRAM + RAM) to be in the range of 80GB or so in my tests for at least 5 tokens or so /s.
It's better to get (VRAM + RAM) >= 140GB for at least 30 to 40 tokens/s, and if VRAM >= 140GB, then it can approach 140 tokens/s!
Another trick is to accept more than 8 experts per pass - it'll be slower, but might be more accurate. You could even try reducing the # of experts to say 6 or 7 for low FLOP machines!
It's better to get (VRAM + RAM) >= 140GB for at least 30 to 40 tokens/s, and if VRAM >= 140GB, then it can approach 140 tokens/s!
Another trick is to accept more than 8 experts per pass - it'll be slower, but might be more accurate. You could even try reducing the # of experts to say 6 or 7 for low FLOP machines!