gpu4.pics
data via LocalScore
NVIDIA GeForce RTX 3070 Ti
8 GB VRAM NVIDIA
running Llama 3.2 1B Instruct· Q4_K_M· 1.5B
297
tokens / sec
2,795 · Excellent #27 of 768
Generation
297tok/s
higher is better
Prompt / prefill
10,074tok/s
higher is better
Time to first token
137 ms
lower is better
Composite score
2,795
Model size
1.5B
Runtime
gpu4.pics/r/ls-nvidia-geforce-rtx-3070-ti-llama-32-1b-instruct-q4-k-m

All benchmarks on this rig

ModelQuantGenPromptTTFTPowerPerf/W$/MtokScore
Llama 3.2 1B Instruct1.5BQ4_K_M29710,074137 ms2,795

How it stacks up — Llama 3.2 1B Instruct

NVIDIA GeForce RTX 3070 Ti
297
NVIDIA L40S
355
NVIDIA RTX 3090 Ti
354
NVIDIA H100
335
NVIDIA RTX 3090
330
NVIDIA RTX A6000
315
NVIDIA A100 80GB
308