← Back to community benchmarks

Qwen3.5-122B-A10B-oQ5

M2 Ultra (76c) · 128 GB · 5bit · 2026-03-26
Performance
1k
tokens
403.8
PP tok/s
29.9
TG tok/s
2536
TTFT (ms)
82.1
Peak mem (GB)
Hardware
Chip M2 Ultra (76c)
Memory 128 GB
GPU Cores 76
Software
oMLX v0.2.21
macOS macOS 26.3.1
Context 1,024
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 403.8 29.9 82.1 GB current
4k 437.8 27.6 83.5 GB view
8k 436.6 26.0 84.1 GB view
16k 439.7 24.8 85.3 GB view