← Back to community benchmarks

Qwen3.6-35B-A3B-MXFP4-MTP

M3 Max (40c) · 48 GB · mxfp4 · 2026-05-20
Performance
4k
tokens
1,610
PP tok/s
54.9
TG tok/s
2544
TTFT (ms)
22.6
Peak mem (GB)
Hardware
Chip M3 Max (40c)
Memory 48 GB
GPU Cores 40
Software
oMLX v0.3.9.dev2
macOS macOS 26.5
Context 4,096
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 1,310 57.0 21.8 GB view
4k 1,610 54.9 22.6 GB current
Batching Results
Batch Size TG tok/s Speedup
57.0 1.00×
156.3 2.74×
236.2 4.14×