GPUs / Qualcomm

Adreno 830

Memory
LPDDR5X-10600
Bus
Not published
Published peak
Not published
Capacity
from 12 GB

Measured

Read
64.7GB/s
Write
55.9GB/s
Copy
61.1GB/s
Random
0.44GB/s
Latency
313ns
Sustain
63.5GB/s

Medians of 1 verified run from 1 operator.

Local AI speed

ModelParameters4-bit, tokens/s8-bit, tokens/s
Llama 3.1 8B8.03B≤ 16≤ 8.1
Qwen 2.5 14B14.7B≤ 8.8Does not fit
gpt-oss-20b21B, 3.6B active≤ 36Does not fit
Mistral Small 24B24B≤ 5.4Does 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 12 GB configuration; the KV cache needs room on top.

Top operators

#OperatorScoreReadDate
1OP-FC9E6930.264.7 GB/s2026-09-27

Full Adreno 830 leaderboard