GPUs / NVIDIA

GTX 1650

Memory
GDDR5
Bus
128-bit
Published peak
128 to 192 GB/s
Capacity
from 4 GB

Measured

Read
111.8GB/s
Write
104.0GB/s
Copy
104.0GB/s
Random
8.62GB/s
Latency
294ns
Sustain
106.4GB/s

Medians of 1 verified run from 1 operator.

Local AI speed

ModelParameters4-bit, tokens/s8-bit, tokens/s
Llama 3.1 8B8.03BDoes not fitDoes not fit
Qwen 2.5 14B14.7BDoes not fitDoes not fit
gpt-oss-20b21B, 3.6B activeDoes not fitDoes not fit
Mistral Small 24B24BDoes not fitDoes not fit
Qwen 2.5 32B32.5BDoes not fitDoes not fit
Llama 3.3 70B70.6BDoes not fitDoes not fit
gpt-oss-120b117B, 5.1B activeDoes not fitDoes not fit
Upper bounds: median read ÷ bytes read per token. Mixture-of-experts models read only their active experts. Fit checks weights against the smallest 4 GB configuration; the KV cache needs room on top.

Top operators

#OperatorScoreReadDate
1OP-96E16F49.5111.8 GB/s2026-09-27

Full GTX 1650 leaderboard