← Back to community benchmarks
Qwen3.6-27B-UD
Performance
8k
tokens
50.3
PP tok/s
5.7
TG tok/s
162816
TTFT (ms)
27.3
Peak mem (GB)
Hardware
Chip
M1 Max (32c)
Memory
64 GB
GPU Cores
32
Software
oMLX
v0.3.7
macOS
macOS 26.4.1
Context
8,192
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 51.7 | 7.0 | 25.2 GB | view |
| 4k | 52.7 | 6.1 | 26.6 GB | view |
| 8k | 50.3 | 5.7 | 27.3 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 7.0 | 1.00× |
| 2× | 5.9 | 0.84× |
| 4× | 8.2 | 1.17× |
| 8× | 8.1 | 1.16× |