Llama 3.2 3B Instruct
Q4_K - Medium
3.6Bparams
COMPARE ACCELERATORS
11 accelerators tested
Select Accelerators
NVIDIA GeForce RTX 5060 Ti
16GB
NVIDIA GeForce RTX 4050 Laptop GPU
6GB
AMD Radeon RX 6900 XT
16GB
NVIDIA GeForce GTX 1060 6GB
6GB
AMD Ryzen 9 5950X 16-Core Processor (znver3)
125.7GB
Llama 3.2 3B Instruct - Q4_K - Medium
LEADERBOARD
GPU / 6GB
PROMPT
2518
tokens/s
GENERATION
60.0
tokens/s
TTFT
568
ms
LOCALSCORE
643
GPU / 6GB
PROMPT
293
tokens/s
GENERATION
14.4
tokens/s
TTFT
4.99
sec
LOCALSCORE
95
