← Back to community benchmarks

Qwen3.5-0.8B

M1 (7c) · 16 GB · 8bit · 2026-03-09
Performance
1k
tokens
719.3
PP tok/s
56.2
TG tok/s
1424
TTFT (ms)
1.8
Peak mem (GB)
Hardware
Chip M1 (7c)
Memory 16 GB
GPU Cores 7
Software
oMLX v0.2.6
macOS macOS 26.3.1
Context 1,024
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 719.3 56.2 1.8 GB current
4k 730.4 52.8 1.9 GB view
Batching Results
Batch Size TG tok/s Speedup
56.2 1.00×
117.0 2.08×
136.7 2.43×