← Back to community benchmarks
Qwen3.6-35B-A3B
Performance
16k
tokens
1,611
PP tok/s
41.2
TG tok/s
10168
TTFT (ms)
67.4
Peak mem (GB)
Hardware
Chip
M1 Ultra (64c)
Memory
128 GB
GPU Cores
64
Software
oMLX
v0.3.6
macOS
macOS 26.5
Context
16,384
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 1,074 | 47.4 | 65.6 GB | view |
| 4k | 1,618 | 46.9 | 66.4 GB | view |
| 8k | 1,683 | 45.8 | 66.7 GB | view |
| 16k | 1,611 | 41.2 | 67.4 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 47.4 | 1.00× |
| 2× | 50.8 | 1.07× |
| 4× | 87.3 | 1.84× |