← Back to community benchmarks
Qwen3.6-35B-A3B
Performance
64k
tokens
248.0
PP tok/s
17.8
TG tok/s
264207
TTFT (ms)
24.9
Peak mem (GB)
Hardware
Chip
M4 (10c)
Memory
32 GB
GPU Cores
10
Software
oMLX
v0.3.5.dev1
macOS
macOS 26.3.1
Context
65,536
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 422.8 | 46.9 | 19.2 GB | view |
| 4k | 461.1 | 44.5 | 19.9 GB | view |
| 8k | 446.6 | 41.3 | 20.2 GB | view |
| 16k | 410.3 | 35.7 | 20.9 GB | view |
| 32k | 343.2 | 27.3 | 22.2 GB | view |
| 64k | 248.0 | 17.8 | 24.9 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 46.9 | 1.00× |
| 2× | 68.5 | 1.46× |
| 4× | 83.8 | 1.79× |