← Back to community benchmarks

DeepSeek-V3.25bit

M3 Ultra (80c) · 512 GB · 5bit · 2026-03-14
Performance
1k
tokens
190.6
PP tok/s
16.2
TG tok/s
5371
TTFT (ms)
432.4
Peak mem (GB)
Hardware
Chip M3 Ultra (80c)
Memory 512 GB
GPU Cores 80
Software
oMLX v0.2.12
macOS macOS 15.5
Context 1,024
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 190.6 16.2 432.4 GB current
4k 186.2 13.6 434.8 GB view
8k 161.0 13.1 438.3 GB view