← Back to community benchmarks

Qwen3.6-35B-A3B-Holo3-Qwopus

M5 (10c) · 32 GB · 4bit · 2026-04-25
Performance
4k
tokens
1,164
PP tok/s
59.4
TG tok/s
3520
TTFT (ms)
18.9
Peak mem (GB)
Hardware
Chip M5 (10c)
Memory 32 GB
GPU Cores 10
Software
oMLX v0.3.6
macOS macOS 26.3.1
Context 4,096
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 921.8 62.3 18.2 GB view
4k 1,164 59.4 18.9 GB current
Batching Results
Batch Size TG tok/s Speedup
62.3 1.00×
93.2 1.50×
107.6 1.73×