← Back to community benchmarks

Step-3.5-Flash-REAP-149B-A11B-4bitCRACK

M3 Ultra (60c) · 256 GB · 4bit · 2026-03-18
Performance
32k
tokens
445.2
PP tok/s
23.3
TG tok/s
73602
TTFT (ms)
80.6
Peak mem (GB)
Hardware
Chip M3 Ultra (60c)
Memory 256 GB
GPU Cores 60
Software
oMLX v0.2.17
macOS macOS 26.2
Context 32,768
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 486.9 54.0 79.1 GB view
4k 509.2 50.1 79.2 GB view
8k 506.6 42.5 79.4 GB view
16k 486.8 32.6 79.8 GB view
32k 445.2 23.3 80.6 GB current
Batching Results
Batch Size TG tok/s Speedup
54.0 1.00×
82.5 1.53×
106.9 1.98×
131.3 2.43×