← Back to community benchmarks

Huihui-Qwen3.5-35B-A3B-Claude-4.6-Opus-abliterated

M5 Max (40c) · 128 GB · 4bit · 2026-04-14
Performance
1k
tokens
2,332
PP tok/s
133.0
TG tok/s
439
TTFT (ms)
19.3
Peak mem (GB)
Hardware
Chip M5 Max (40c)
Memory 128 GB
GPU Cores 40
Software
oMLX v0.3.5
macOS macOS 26.4
Context 1,024
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 2,332 133.0 19.3 GB current
4k 3,901 128.0 20.0 GB view
Batching Results
Batch Size TG tok/s Speedup
133.0 1.00×
222.6 1.67×
332.8 2.50×