← Back to community benchmarks

Qwen3.5-0.8B

M3 (10c) · 16 GB · 8bit · 2026-03-09
Performance
1k
tokens
1,877
PP tok/s
97.0
TG tok/s
546
TTFT (ms)
1.8
Peak mem (GB)
Hardware
Chip M3 (10c)
Memory 16 GB
GPU Cores 10
Software
oMLX v0.2.6
macOS macOS 26.2
Context 1,024
Batching Results
Batch Size TG tok/s Speedup
1× 97.0 1.00×
2× 179.0 1.85×
4× 282.4 2.91×
8× 293.2 3.02×