← Back to community benchmarks

Qwen3.5-27B-Claude-4.6-Opus-Distilled

M4 (10c) · 32 GB · 4bit · 2026-03-12
User commentQwen3.5-27B-Claude-4.6-Opus-Distilled · 4bit
Performance
4k
tokens
59.2
PP tok/s
6.2
TG tok/s
69204
TTFT (ms)
16.1
Peak mem (GB)
Hardware
Chip M4 (10c)
Memory 32 GB
GPU Cores 10
Software
oMLX v0.2.8
macOS macOS 26.3.1
Context 4,096
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 61.4 6.5 15.7 GB view
4k 59.2 6.2 16.1 GB current
Batching Results
Batch Size TG tok/s Speedup
6.5 1.00×
13.1 2.02×
15.8 2.43×
16.3 2.51×