← Back to community benchmarks
Step-3.5-Flash6.5bit
Performance
16k
tokens
262.6
PP tok/s
24.4
TG tok/s
62395
TTFT (ms)
153.2
Peak mem (GB)
Hardware
Chip
M2 Ultra (76c)
Memory
192 GB
GPU Cores
76
Software
oMLX
v0.2.6
macOS
macOS 26.3
Context
16,384
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 313.4 | 35.5 | 150.0 GB | view |
| 8k | 307.3 | 29.4 | 151.7 GB | view |
| 16k | 262.6 | 24.4 | 153.2 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 35.5 | 1.00× |
| 2× | 50.4 | 1.42× |
| 4× | 63.5 | 1.79× |