← Back to community benchmarks

Qwen3.6-35B-A3B

M1 Ultra (64c) · 128 GB · fp16 · 2026-04-17
Performance
16k
tokens
1,895
PP tok/s
44.6
TG tok/s
8644
TTFT (ms)
67.4
Peak mem (GB)
Hardware
Chip M1 Ultra (64c)
Memory 128 GB
GPU Cores 64
Software
oMLX v0.3.6
macOS macOS 26.5
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 1,036 50.6 65.6 GB view
4k 1,752 48.4 66.4 GB view
8k 1,953 47.8 66.7 GB view
16k 1,895 44.6 67.4 GB current
Batching Results
Batch Size TG tok/s Speedup
50.6 1.00×
56.3 1.11×
94.6 1.87×