← Back to community benchmarks

Qwen3.6-35B-A3BoQ8

M5 Max (40c) · 128 GB · 8bit · 2026-05-01
Performance
64k
tokens
2,357
PP tok/s
49.3
TG tok/s
27808
TTFT (ms)
41.1
Peak mem (GB)
Hardware
Chip M5 Max (40c)
Memory 128 GB
GPU Cores 40
Software
oMLX v0.3.8rc1
macOS macOS 26.5
Context 65,536
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
64k 2,357 49.3 41.1 GB current
128k 1,385 38.4 46.5 GB view