← Back to community benchmarks

Qwen3.5-0.8B

M4 (10c) · 16 GB · unknown · 2026-03-11
Performance
16k
tokens
1,470
PP tok/s
12.7
TG tok/s
11142
TTFT (ms)
4.4
Peak mem (GB)
Hardware
Chip M4 (10c)
Memory 16 GB
GPU Cores 10
Software
oMLX v0.2.7
macOS macOS 26.3.1
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 1,419 13.6 3.3 GB view
4k 1,677 13.4 3.4 GB view
8k 1,627 13.2 3.6 GB view
16k 1,470 12.7 4.4 GB current
Batching Results
Batch Size TG tok/s Speedup
13.6 1.00×
26.1 1.92×
50.6 3.72×
98.1 7.21×