← Back to community benchmarks

Qwen3.5-2B-Polaris-HighIQ-INSTRUCT-oQ8

M4 Pro (16c) · 24 GB · 8bit · 2026-04-15
Performance
1k
tokens
1,226
PP tok/s
105.8
TG tok/s
835
TTFT (ms)
2.8
Peak mem (GB)
Hardware
Chip M4 Pro (16c)
Memory 24 GB
GPU Cores 16
Software
oMLX v0.3.5
macOS macOS 26.4.1
Context 1,024
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 1,226 105.8 2.8 GB current
4k 1,510 95.9 3.2 GB view
Batching Results
Batch Size TG tok/s Speedup
105.8 1.00×
180.2 1.70×
221.9 2.10×