← Back to community benchmarks
Qwen3.6-35B-A3B
Performance
16k
tokens
3,793
PP tok/s
87.9
TG tok/s
4320
TTFT (ms)
37.1
Peak mem (GB)
Hardware
Chip
M5 Max (40c)
Memory
128 GB
GPU Cores
40
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 | 2,199 | 100.9 | 35.4 GB | view |
| 4k | 3,663 | 97.6 | 36.2 GB | view |
| 8k | 3,907 | 93.9 | 36.5 GB | view |
| 16k | 3,793 | 87.9 | 37.1 GB | current |
| 32k | 3,194 | 77.3 | 38.5 GB | view |
| 64k | 2,231 | 61.5 | 41.2 GB | view |
| 128k | 1,365 | 33.1 | 46.5 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 100.9 | 1.00× |
| 2× | 159.6 | 1.58× |
| 4× | 233.2 | 2.31× |