← Back to community benchmarks
Qwen3.6-35B-A3B
Performance
16k
tokens
954.9
PP tok/s
66.2
TG tok/s
17158
TTFT (ms)
36.7
Peak mem (GB)
Hardware
Chip
M2 Ultra (60c)
Memory
64 GB
GPU Cores
60
Software
oMLX
v0.5.1
macOS
macOS 27.0
Context
16,384
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 1,081 | 70.4 | 35.4 GB | view |
| 4k | 1,096 | 59.4 | 36.1 GB | view |
| 8k | 1,016 | 36.2 | 36.4 GB | view |
| 16k | 954.9 | 66.2 | 36.7 GB | current |
| 32k | 775.4 | 60.5 | 37.4 GB | view |
| 64k | 562.4 | 49.1 | 38.5 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 70.4 | 1.00× |
| 2× | 115.0 | 1.63× |
| 4× | 109.0 | 1.55× |