← Back to community benchmarks
Qwen3.6-35B-A3B
Performance
16k
tokens
1,121
PP tok/s
56.8
TG tok/s
14616
TTFT (ms)
37.1
Peak mem (GB)
Hardware
Chip
M2 Ultra (60c)
Memory
64 GB
GPU Cores
60
Software
oMLX
v0.3.6
macOS
macOS 26.4.1
Context
16,384
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 860.4 | 71.5 | 35.4 GB | view |
| 4k | 1,118 | 68.0 | 36.2 GB | view |
| 8k | 1,156 | 66.1 | 36.5 GB | view |
| 16k | 1,121 | 56.8 | 37.1 GB | current |
| 32k | 1,022 | 42.2 | 38.5 GB | view |
| 64k | 814.7 | 29.5 | 41.2 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 71.5 | 1.00× |
| 2× | 121.0 | 1.69× |
| 4× | 184.0 | 2.57× |
| 8× | 264.1 | 3.69× |