← Back to community benchmarks
Qwen3___5-4B
Performance
32k
tokens
2,116
PP tok/s
90.5
TG tok/s
15484
TTFT (ms)
6.9
Peak mem (GB)
Hardware
Chip
M3 Ultra (80c)
Memory
512 GB
GPU Cores
80
Software
oMLX
v0.2.6
macOS
macOS 26.3.1
Context
32,768
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 2,581 | 152.6 | 4.0 GB | view |
| 4k | 2,608 | 143.4 | 4.2 GB | view |
| 8k | 2,542 | 133.1 | 4.6 GB | view |
| 16k | 2,385 | 116.4 | 5.3 GB | view |
| 32k | 2,116 | 90.5 | 6.9 GB | current |
| 64k | 1,701 | 61.3 | 8.6 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 152.6 | 1.00× |
| 2× | 263.9 | 1.73× |
| 4× | 371.8 | 2.44× |