← Back to community benchmarks
Qwen3.5-9B
Performance
8k
tokens
1,443
PP tok/s
59.7
TG tok/s
5677
TTFT (ms)
11.2
Peak mem (GB)
Hardware
Chip
M3 Ultra (80c)
Memory
512 GB
GPU Cores
80
Software
oMLX
v0.2.10
macOS
macOS 15.3
Context
8,192
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 1,424 | 64.1 | 10.7 GB | view |
| 4k | 1,457 | 62.1 | 11.0 GB | view |
| 8k | 1,443 | 59.7 | 11.2 GB | current |
| 16k | 1,395 | 54.7 | 11.7 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 64.1 | 1.00× |
| 2× | 112.0 | 1.75× |
| 4× | 166.8 | 2.60× |
| 8× | 210.1 | 3.28× |