← Back to community benchmarks
Qwen3.6-27B-MXFP4-MTP
Performance
16k
tokens
167.2
PP tok/s
7.1
TG tok/s
97997
TTFT (ms)
18.3
Peak mem (GB)
Hardware
Chip
M5 (10c)
Memory
32 GB
GPU Cores
10
Software
oMLX
v0.3.6
macOS
macOS 26.4
Context
16,384
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 193.4 | 8.6 | 15.0 GB | view |
| 4k | 199.6 | 8.6 | 16.4 GB | view |
| 8k | 191.0 | 7.9 | 17.0 GB | view |
| 16k | 167.2 | 7.1 | 18.3 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 8.6 | 1.00× |
| 2× | 15.1 | 1.76× |
| 4× | 86.5 | 10.06× |