← Back to community benchmarks
Qwen3.6-35B-A3B-UD-Q4_K_XL
Performance
1k
tokens
2,226
PP tok/s
104.0
TG tok/s
460
TTFT (ms)
21.5
Peak mem (GB)
Hardware
Chip
M5 Max (40c)
Memory
128 GB
GPU Cores
40
Software
oMLX
v0.3.8
macOS
macOS 26.4
Context
1,024
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 2,226 | 104.0 | 21.5 GB | current |
| 4k | 3,732 | 102.2 | 22.3 GB | view |
| 8k | 3,991 | 101.0 | 22.8 GB | view |
| 16k | 3,588 | 95.8 | 23.5 GB | view |
| 32k | 2,908 | 88.4 | 25.0 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 104.0 | 1.00× |
| 2× | 125.5 | 1.21× |
| 4× | 201.0 | 1.93× |
| 8× | 491.0 | 4.72× |