← Back to community benchmarks

Harmonic-Hermes-9B

M5 Max (40c) · 128 GB · 8bit · 2026-04-07
Performance
1k
tokens
1,979
PP tok/s
59.0
TG tok/s
517
TTFT (ms)
9.9
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 1,979 59.0 9.9 GB current
4k 2,779 57.6 10.6 GB view
8k 2,840 55.9 11.0 GB view
Batching Results
Batch Size TG tok/s Speedup
59.0 1.00×
110.8 1.88×
150.8 2.56×