logo
HomeLatest ResultsDownloadAboutBlog
model icon

Llama 3 8b Bnb 4bit

Q4_K - Medium

8.0Bparams
COMPARE ACCELERATORS

1 accelerators tested

Select Accelerators

NVIDIA GeForce RTX 3080

10GB

Llama 3 8b Bnb 4bit - Q4_K - Medium

LEADERBOARD
NVIDIA GeForce RTX 3080
GPU / 10GB
PROMPT
3204
tokens/s
GENERATION
78.3
tokens/s
TTFT
403
ms
LOCALSCORE
854