← Back to community benchmarks
Qwen3-30B-A3B
Performance
16k
tokens
639.6
PP tok/s
26.3
TG tok/s
25616
TTFT (ms)
18.1
Peak mem (GB)
Hardware
Chip
M5 (10c)
Memory
32 GB
GPU Cores
10
Software
oMLX
v0.2.11
macOS
macOS 26.3.1
Context
16,384
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 864.3 | 52.6 | 16.6 GB | view |
| 4k | 993.4 | 43.8 | 17.0 GB | view |
| 8k | 877.7 | 36.6 | 17.4 GB | view |
| 16k | 639.6 | 26.3 | 18.1 GB | current |
| 32k | 398.4 | 16.7 | 19.6 GB | view |
| 64k | 221.9 | 10.0 | 22.7 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 52.6 | 1.00× |
| 2× | 70.5 | 1.34× |
| 4× | 82.3 | 1.56× |