← Back to community benchmarks

Qwen3.5-35B-A3B-Claude-4.6-Opus-Reasoning-Distilled

M3 Max (40c) · 48 GB · 4bit · 2026-05-15
Performance
4k
tokens
1,236
PP tok/s
61.9
TG tok/s
3314
TTFT (ms)
20.0
Peak mem (GB)
Hardware
Chip M3 Max (40c)
Memory 48 GB
GPU Cores 40
Software
oMLX v0.3.9.dev1
macOS macOS 26.5
Context 4,096
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 899.1 66.4 19.3 GB view
4k 1,236 61.9 20.0 GB current
8k 1,060 67.9 20.5 GB view
16k 995.2 61.4 21.2 GB view
Batching Results
Batch Size TG tok/s Speedup
66.4 1.00×
123.3 1.86×
180.5 2.72×
211.5 3.19×