top | item 38933788 (no title) teilo | 2 years ago No quantization (8_0). The full 48GB model. As for token count, I haven't tested it on more than 200 or so. discuss order hn newest pilotneko|2 years ago Isn’t 8_0 8-bit quantization? teilo|2 years ago Sorry. That was a major brain fart. Yes. 8-bit quantization, and using 49G of RAM.
pilotneko|2 years ago Isn’t 8_0 8-bit quantization? teilo|2 years ago Sorry. That was a major brain fart. Yes. 8-bit quantization, and using 49G of RAM.
pilotneko|2 years ago
teilo|2 years ago