← Back to community benchmarks

Qwen3.5-0.8B

M5 (10c) · 24 GB · 8bit · 2026-03-09
Performance
1k
tokens
2,302
PP tok/s
132.1
TG tok/s
445
TTFT (ms)
1.8
Peak mem (GB)
Hardware
Chip M5 (10c)
Memory 24 GB
GPU Cores 10
Software
oMLX v0.2.6
macOS macOS 26.3
Context 1,024
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 2,302 132.1 1.8 GB current
4k 2,382 130.0 1.9 GB view
Batching Results
Batch Size TG tok/s Speedup
132.1 1.00×
270.4 2.05×
443.5 3.36×