← Back to community benchmarks

Qwen3.6-35B-A3B

M3 Max (40c) · 48 GB · 8bit · 2026-04-24
Performance
4k
tokens
1,397
PP tok/s
65.2
TG tok/s
2933
TTFT (ms)
36.1
Peak mem (GB)
Hardware
Chip M3 Max (40c)
Memory 48 GB
GPU Cores 40
Software
oMLX v0.3.7
macOS macOS 26.3.1
Context 4,096
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 1,014 65.5 35.4 GB view
4k 1,397 65.2 36.1 GB current
Batching Results
Batch Size TG tok/s Speedup
65.5 1.00×
110.9 1.69×
168.5 2.57×