← Back to community benchmarks
Qwen3.6-35B-A3B
Performance
128k
tokens
1,036
PP tok/s
35.3
TG tok/s
126503
TTFT (ms)
30.4
Peak mem (GB)
Hardware
Chip
M3 Ultra (80c)
Memory
512 GB
GPU Cores
80
Software
oMLX
v0.3.4
macOS
macOS 26.3.1
Context
131,072
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 1,688 | 98.7 | 19.2 GB | view |
| 8k | 2,671 | 88.7 | 20.4 GB | view |
| 32k | 2,203 | 69.2 | 22.4 GB | view |
| 64k | 1,662 | 52.8 | 25.0 GB | view |
| 128k | 1,036 | 35.3 | 30.4 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 98.7 | 1.00× |
| 2× | 172.6 | 1.75× |
| 4× | 351.2 | 3.56× |
| 8× | 678.2 | 6.87× |