← Back to community benchmarks
Qwen3.6-27B-oQ4e
Performance
16k
tokens
160.5
PP tok/s
20.6
TG tok/s
102098
TTFT (ms)
20.3
Peak mem (GB)
Hardware
Chip
M4 Max (40c)
Memory
128 GB
GPU Cores
40
Software
oMLX
v0.3.8
macOS
macOS 15.7.5
Context
16,384
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 224.9 | 26.3 | 16.3 GB | view |
| 4k | 203.7 | 22.8 | 17.7 GB | view |
| 8k | 171.0 | 17.6 | 18.8 GB | view |
| 16k | 160.5 | 20.6 | 20.3 GB | current |
| 32k | 164.0 | 18.7 | 23.3 GB | view |
| 64k | 142.7 | 16.4 | 29.3 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 26.3 | 1.00× |
| 2× | 35.8 | 1.36× |
| 4× | 43.6 | 1.66× |
| 8× | 43.1 | 1.64× |