← Back to community benchmarks
Qwen3.6-35B-A3B-oQ4
Performance
64k
tokens
2,319
PP tok/s
69.1
TG tok/s
28261
TTFT (ms)
25.7
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
65,536
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 2,257 | 127.3 | 19.9 GB | view |
| 4k | 3,802 | 115.8 | 20.7 GB | view |
| 8k | 4,070 | 114.0 | 21.1 GB | view |
| 16k | 3,919 | 105.3 | 21.7 GB | view |
| 32k | 3,347 | 90.9 | 23.0 GB | view |
| 64k | 2,319 | 69.1 | 25.7 GB | current |
| 128k | 1,343 | 47.0 | 31.1 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 127.3 | 1.00× |
| 2× | 201.5 | 1.58× |
| 4× | 283.8 | 2.23× |