← Back to community benchmarks

Qwen3.6-27B-oQ8-mtp

M1 Max (32c) · 64 GB · 8bit · 2026-06-03
User commentmtp disabled
Performance
1k
tokens
75.4
PP tok/s
10.3
TG tok/s
13580
TTFT (ms)
28.3
Peak mem (GB)
Hardware
Chip M1 Max (32c)
Memory 64 GB
GPU Cores 32
Software
oMLX v0.3.9.dev1
macOS macOS 26.4
Context 1,024
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 75.4 10.3 28.3 GB current
4k 59.2 8.8 29.8 GB view
Batching Results
Batch Size TG tok/s Speedup
10.3 1.00×
9.0 0.87×
11.8 1.15×