← Back to community benchmarks

supergemma4-26b-uncensored4bit-v2

M1 Max (24c) · 64 GB · 4bit · 2026-05-04
Performance
16k
tokens
367.6
PP tok/s
39.7
TG tok/s
44567
TTFT (ms)
15.3
Peak mem (GB)
Hardware
Chip M1 Max (24c)
Memory 64 GB
GPU Cores 24
Software
oMLX v0.3.8
macOS macOS 26.3
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 347.9 50.4 14.0 GB view
4k 373.4 47.5 14.7 GB view
8k 376.8 44.5 14.8 GB view
16k 367.6 39.7 15.3 GB current
32k 344.5 32.6 16.2 GB view
64k 303.5 20.9 18.3 GB view
Batching Results
Batch Size TG tok/s Speedup
50.4 1.00×
60.9 1.21×
70.9 1.41×