gpu4.pics
data via LocalScore
NVIDIA RTX 4080
16 GB VRAM NVIDIA 320 W TDP
running Meta Llama 3.1 8B Instruct· Q4_K_M
78.7
tokens / sec
1,136 · Excellent #8 of 370
Generation
78.7tok/s
higher is better
Prompt / prefill
4,979tok/s
higher is better
Time to first token
267 ms
lower is better
Composite score
1,136
Model size
8B
Runtime
gpu4.pics/r/ls-nvidia-rtx-4080-meta-llama-31-8b-instruct-q4-k-m

All benchmarks on this rig

ModelQuantGenPromptTTFTPowerPerf/W$/MtokScore
Meta Llama 3.1 8B InstructQ4_K_M78.74,979267 ms1,136

How it stacks up — Meta Llama 3.1 8B Instruct

NVIDIA RTX 4080
78.7
NVIDIA H100
120
NVIDIA A100 80GB
110
NVIDIA RTX 3090 Ti
110
NVIDIA RTX PRO 6000 Blackwell Workstation Edition
103
NVIDIA RTX 3090
103
NVIDIA RTX A6000
90.5