← Back to community benchmarks

Qwen2.5-7B-Instruct

M1 (8c) · 16 GB · 4bit · 2026-03-11
Performance
1k
tokens
142.4
PP tok/s
14.3
TG tok/s
7189
TTFT (ms)
4.5
Peak mem (GB)
Hardware
Chip M1 (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 142.4 14.3 4.5 GB current
4k 138.3 10.5 4.8 GB view
Batching Results
Batch Size TG tok/s Speedup
14.3 1.00×
19.2 1.34×
23.3 1.63×
29.6 2.07×