← Back to community benchmarks

Qwen3-4B-Gemini-Flash-Distilled-Instruct-qx86-hi

M3 (8c) · 16 GB · 8bit · 2026-03-11
Performance
1k
tokens
257.4
PP tok/s
22.5
TG tok/s
3978
TTFT (ms)
4.1
Peak mem (GB)
Hardware
Chip M3 (8c)
Memory 16 GB
GPU Cores 8
Software
oMLX v0.2.7
macOS macOS 26.3.1
Context 1,024
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 257.4 22.5 4.1 GB current
4k 234.3 19.4 4.6 GB view
Batching Results
Batch Size TG tok/s Speedup
22.5 1.00×
36.5 1.62×
39.0 1.73×