3 ms·
Llama 3.2 punches way above its weight. For general "language manipulation" tasks it's good enough - and it can be used on a CPU with acceptable speed.
by RicoElectrico 2y ago
Llama 3.2 punches way above its weight. For general "language manipulation" tasks it's good enough - and it can be used on a CPU with acceptable speed.
- seunosewa 2y agoHow many tokens/s?
- iamnotagenius 2y ago10-15t/s on 12400 with ddr5