gpu4.pics
data via LocalScore
NVIDIA RTX 4090
24 GB VRAM NVIDIA 450 W TDP
running Llama 3.2 1B Instruct· Q4_K_M· 1.5B
207
tokens / sec
2,559 · Excellent #36 of 768
Generation
207tok/s
higher is better
Prompt / prefill
11,965tok/s
higher is better
Time to first token
148 ms
lower is better
Composite score
2,559
Model size
1.5B
Runtime
gpu4.pics/r/ls-nvidia-rtx-4090-llama-32-1b-instruct-q4-k-m

All benchmarks on this rig

ModelQuantGenPromptTTFTPowerPerf/W$/MtokScore
Llama 3.2 1B Instruct1.5BQ4_K_M20711,965148 ms2,559

How it stacks up — Llama 3.2 1B Instruct

NVIDIA RTX 4090
207
NVIDIA L40S
355
NVIDIA RTX 3090 Ti
354
NVIDIA H100
335
NVIDIA RTX 3090
330
NVIDIA RTX A6000
315
NVIDIA A100 80GB
308