← Back to community benchmarks

Qwen3.5-0.8B

M2 (10c) · 8 GB · 8bit · 2026-03-19
Performance
4k
tokens
1,182
PP tok/s
82.0
TG tok/s
3465
TTFT (ms)
2.3
Peak mem (GB)
Hardware
Chip M2 (10c)
Memory 8 GB
GPU Cores 10
Software
oMLX v0.2.19
macOS macOS 26.3.1
Context 4,096
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 1,179 88.0 1.9 GB view
4k 1,182 82.0 2.3 GB current
Batching Results
Batch Size TG tok/s Speedup
88.0 1.00×
159.7 1.81×
215.8 2.45×
215.6 2.45×