← Back to community benchmarks
Qwen3.5-9B-DeepSeek-V4-Flash-oQ6
Performance
4k
tokens
2,677
PP tok/s
69.1
TG tok/s
1530
TTFT (ms)
8.8
Peak mem (GB)
Hardware
Chip
M5 Max (40c)
Memory
128 GB
GPU Cores
40
Software
oMLX
v0.3.8
macOS
macOS 26.4.1
Context
4,096
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 1,856 | 70.7 | 8.0 GB | view |
| 4k | 2,677 | 69.1 | 8.8 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 70.7 | 1.00× |
| 2× | 121.9 | 1.72× |
| 4× | 167.3 | 2.37× |