← Back to community benchmarks

Harmonic-Hermes-9B

M5 Max (40c) · 128 GB · bf16 · 2026-04-07
Performance
1k
tokens
1,764
PP tok/s
31.1
TG tok/s
581
TTFT (ms)
17.5
Peak mem (GB)
Hardware
Chip M5 Max (40c)
Memory 128 GB
GPU Cores 40
Software
oMLX v0.3.4
macOS macOS 26.5
Context 1,024
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 1,764 31.1 17.5 GB current
4k 2,930 30.6 18.6 GB view
8k 2,907 30.2 18.9 GB view
Batching Results
Batch Size TG tok/s Speedup
31.1 1.00×
36.6 1.18×
68.5 2.20×