← Back to community benchmarks

Qwen3.6-35B-A3B-oQ8

M1 Ultra (64c) · 128 GB · 8bit · 2026-04-17
Performance
1k
tokens
1,082
PP tok/s
57.6
TG tok/s
947
TTFT (ms)
35.4
Peak mem (GB)
Hardware
Chip M1 Ultra (64c)
Memory 128 GB
GPU Cores 64
Software
oMLX v0.3.6
macOS macOS 26.5
Context 1,024
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 1,082 57.6 35.4 GB current
4k 1,715 58.6 36.2 GB view
8k 1,816 55.3 36.5 GB view
16k 1,760 51.3 37.1 GB view
Batching Results
Batch Size TG tok/s Speedup
57.6 1.00×
100.7 1.75×
156.6 2.72×