gpu4.pics
data via LocalScore
NVIDIA GeForce RTX 2070 Super
8 GB VRAM NVIDIA
running Llama 3.2 1B Instruct· Q4_K_M· 1.5B
189
tokens / sec
1,660 · Excellent #74 of 768
Generation
189tok/s
higher is better
Prompt / prefill
5,696tok/s
higher is better
Time to first token
236 ms
lower is better
Composite score
1,660
Model size
1.5B
Runtime
gpu4.pics/r/ls-nvidia-geforce-rtx-2070-super-llama-32-1b-instruct-q4-k-m

All benchmarks on this rig

ModelQuantGenPromptTTFTPowerPerf/W$/MtokScore
Llama 3.2 1B Instruct1.5BQ4_K_M1895,696236 ms1,660

How it stacks up — Llama 3.2 1B Instruct

NVIDIA GeForce RTX 2070 Super
189
NVIDIA L40S
355
NVIDIA RTX 3090 Ti
354
NVIDIA H100
335
NVIDIA RTX 3090
330
NVIDIA RTX A6000
315
NVIDIA A100 80GB
308