← Back to community benchmarks
Qwen3.5-9B-DeepSeek-V4-Flash
Performance
4k
tokens
214.3
PP tok/s
11.2
TG tok/s
19111
TTFT (ms)
10.4
Peak mem (GB)
Hardware
Chip
M4 (10c)
Memory
16 GB
GPU Cores
10
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 | 205.7 | 11.3 | 9.8 GB | view |
| 4k | 214.3 | 11.2 | 10.4 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 11.3 | 1.00× |
| 2× | 22.4 | 1.98× |
| 4× | 35.7 | 3.16× |