← Back to community benchmarks
Qwen3.5-122B-A10B
Performance
8k
tokens
688.5
PP tok/s
38.5
TG tok/s
11899
TTFT (ms)
95.4
Peak mem (GB)
Hardware
Chip
M3 Ultra (60c)
Memory
256 GB
GPU Cores
60
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 | 672.4 | 45.2 | 94.8 GB | view |
| 4k | 701.0 | 43.1 | 95.1 GB | view |
| 8k | 688.5 | 38.5 | 95.4 GB | current |
| 16k | 649.2 | 32.3 | 96.1 GB | view |
| 32k | 572.4 | 22.5 | 97.5 GB | view |
| 64k | 451.9 | 15.1 | 100.3 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 45.2 | 1.00× |
| 2× | 67.8 | 1.50× |
| 4× | 90.0 | 1.99× |
| 8× | 110.1 | 2.44× |