← Back to community benchmarks
Qwen3.6-35B-A3B-oQ4
Performance
128k
tokens
612.4
PP tok/s
26.2
TG tok/s
214042
TTFT (ms)
31.1
Peak mem (GB)
Hardware
Chip
M4 Max (40c)
Memory
64 GB
GPU Cores
40
Software
oMLX
v0.3.5
macOS
macOS 26.4.1
Context
131,072
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 1,361 | 112.1 | 19.9 GB | view |
| 4k | 1,708 | 106.1 | 20.7 GB | view |
| 8k | 1,682 | 99.0 | 21.1 GB | view |
| 16k | 1,572 | 88.5 | 21.7 GB | view |
| 32k | 1,301 | 71.5 | 23.0 GB | view |
| 64k | 988.6 | 40.3 | 25.7 GB | view |
| 128k | 612.4 | 26.2 | 31.1 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 112.1 | 1.00× |
| 2× | 177.7 | 1.59× |
| 4× | 241.2 | 2.15× |