← Back to community benchmarks
Qwen3.6-27B4bit-MTP
Performance
64k
tokens
609.2
PP tok/s
32.6
TG tok/s
107572
TTFT (ms)
29.1
Peak mem (GB)
Hardware
Chip
M5 Max (40c)
Memory
128 GB
GPU Cores
40
Software
oMLX
v0.3.9rc1
macOS
macOS 26.5
Context
65,536
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 732.2 | 45.1 | 16.1 GB | view |
| 4k | 794.5 | 46.0 | 17.5 GB | view |
| 32k | 740.2 | 39.2 | 22.3 GB | view |
| 64k | 609.2 | 32.6 | 29.1 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 45.1 | 1.00× |
| 2× | 57.1 | 1.27× |
| 4× | 72.3 | 1.60× |
| 8× | 79.7 | 1.77× |