← Back to community benchmarks

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

M4 Max (40c) · 128 GB · 4bit · 2026-04-15
Performance
1k
tokens
1,299
PP tok/s
111.0
TG tok/s
788
TTFT (ms)
19.3
Peak mem (GB)
Hardware
Chip M4 Max (40c)
Memory 128 GB
GPU Cores 40
Software
oMLX v0.3.5
macOS macOS 15.4.1
Context 1,024
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 1,299 111.0 19.3 GB current
4k 1,698 106.4 20.0 GB view
Batching Results
Batch Size TG tok/s Speedup
111.0 1.00×
184.9 1.67×
267.1 2.41×