← Back to community benchmarks

Qwen3.5-27B4.5bit

M3 (10c) · 24 GB · 4bit · 2026-04-25
Performance
4k
tokens
53.0
PP tok/s
5.8
TG tok/s
77256
TTFT (ms)
17.2
Peak mem (GB)
Hardware
Chip M3 (10c)
Memory 24 GB
GPU Cores 10
Software
oMLX v0.3.6
macOS macOS 26.4
Context 4,096
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 52.6 5.9 15.8 GB view
4k 53.0 5.8 17.2 GB current
Batching Results
Batch Size TG tok/s Speedup
5.9 1.00×
11.6 1.97×
14.1 2.39×