← Back to community benchmarks

Qwen3.5-122B-A10B-8bit-abliterated

M2 Ultra (76c) · 192 GB · 8bit · 2026-03-20
Performance
32k
tokens
445.7
PP tok/s
14.5
TG tok/s
73517
TTFT (ms)
136.5
Peak mem (GB)
Hardware
Chip M2 Ultra (76c)
Memory 192 GB
GPU Cores 76
Software
oMLX v0.2.19
macOS macOS 26.3.1
Context 32,768
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 597.6 17.8 125.1 GB view
4k 677.9 17.4 128.1 GB view
8k 670.4 17.1 129.3 GB view
16k 556.5 16.0 131.7 GB view
32k 445.7 14.5 136.5 GB current
Batching Results
Batch Size TG tok/s Speedup
17.8 1.00×
31.9 1.79×
51.2 2.88×
79.7 4.48×