← Back to community benchmarks

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

M4 (10c) · 32 GB · 4bit · 2026-04-15
Performance
16k
tokens
389.4
PP tok/s
35.3
TG tok/s
42075
TTFT (ms)
20.9
Peak mem (GB)
Hardware
Chip M4 (10c)
Memory 32 GB
GPU Cores 10
Software
oMLX v0.3.5.dev1
macOS macOS 26.3.1
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 424.7 46.8 19.2 GB view
4k 462.6 44.5 19.9 GB view
8k 439.8 41.1 20.2 GB view
16k 389.4 35.3 20.9 GB current
32k 328.8 27.3 22.2 GB view
Batching Results
Batch Size TG tok/s Speedup
46.8 1.00×
70.5 1.51×