← Back to community benchmarks

Qwen3.5-2B

M1 (8c) · 8 GB · 6bit · 2026-03-10
Performance
1k
tokens
351.3
PP tok/s
31.8
TG tok/s
2915
TTFT (ms)
2.9
Peak mem (GB)
Hardware
Chip M1 (8c)
Memory 8 GB
GPU Cores 8
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 351.3 31.8 2.9 GB current
4k 359.4 31.7 3.0 GB view
Batching Results
Batch Size TG tok/s Speedup
31.8 1.00×
51.7 1.63×
56.3 1.77×