← Back to community benchmarks
Qwen3.6-35B-A3B-oQ8
Performance
16k
tokens
2,459
PP tok/s
82.5
TG tok/s
6663
TTFT (ms)
37.1
Peak mem (GB)
Hardware
Chip
M5 Max (40c)
Memory
128 GB
GPU Cores
40
Software
oMLX
v0.3.5
macOS
macOS 26.4.1
Context
16,384
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 1,583 | 92.0 | 35.4 GB | view |
| 4k | 2,001 | 92.2 | 36.2 GB | view |
| 8k | 2,368 | 89.9 | 36.5 GB | view |
| 16k | 2,459 | 82.5 | 37.1 GB | current |
| 32k | 2,225 | 70.1 | 38.5 GB | view |
| 64k | 1,694 | 52.0 | 41.2 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 92.0 | 1.00× |
| 2× | 146.9 | 1.60× |
| 4× | 196.1 | 2.13× |