← Back to community benchmarks

Qwen3.5-2B9bit

M3 Ultra (60c) · 96 GB · 8bit · 2026-03-09
Performance
16k
tokens
4,253
PP tok/s
163.4
TG tok/s
3852
TTFT (ms)
3.4
Peak mem (GB)
Hardware
Chip M3 Ultra (60c)
Memory 96 GB
GPU Cores 60
Software
oMLX v0.2.6
macOS macOS 26.3.1
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 2,233 200.4 2.9 GB view
4k 3,835 191.8 3.0 GB view
8k 4,214 180.8 3.0 GB view
16k 4,253 163.4 3.4 GB current
32k 3,956 135.9 4.0 GB view
Batching Results
Batch Size TG tok/s Speedup
200.4 1.00×
326.9 1.63×
429.1 2.14×