← Back to community benchmarks
Qwen3.6-27BoQ4
Performance
4k
tokens
207.2
PP tok/s
20.3
TG tok/s
19773
TTFT (ms)
18.1
Peak mem (GB)
Hardware
Chip
M3 Max (40c)
Memory
128 GB
GPU Cores
40
Software
oMLX
v0.3.7
macOS
macOS 26.4.1
Context
4,096
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 208.2 | 21.1 | 16.6 GB | view |
| 4k | 207.2 | 20.3 | 18.1 GB | current |
| 8k | 194.2 | 19.3 | 18.7 GB | view |
| 16k | 189.1 | 16.3 | 19.9 GB | view |
| 32k | 162.1 | 12.1 | 22.4 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 21.1 | 1.00× |
| 2× | 29.6 | 1.40× |
| 4× | 37.4 | 1.77× |