← Back to community benchmarks

Qwopus3.5-122B-A10B-Kimi-K2.6-destill-healed-abliterated-MTPLX

M5 Max (40c) · 128 GB · 4bit · 2026-06-10
Performance
32k
tokens
1,189
PP tok/s
46.6
TG tok/s
27558
TTFT (ms)
71.6
Peak mem (GB)
Hardware
Chip M5 Max (40c)
Memory 128 GB
GPU Cores 40
Software
oMLX v0.4.1
macOS macOS 26.5.1
Context 32,768
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 844.7 58.7 65.6 GB view
4k 1,096 57.0 66.9 GB view
8k 1,215 52.9 67.8 GB view
16k 1,225 51.7 69.0 GB view
32k 1,189 46.6 71.6 GB current
Batching Results
Batch Size TG tok/s Speedup
58.7 1.00×
88.2 1.50×
114.0 1.94×