← Back to community benchmarks
Qwen3.6-27B4bit-MTP
Performance
1k
tokens
87.0
PP tok/s
19.3
TG tok/s
11766
TTFT (ms)
15.8
Peak mem (GB)
Hardware
Chip
M1 Max (32c)
Memory
64 GB
GPU Cores
32
Software
oMLX
v0.3.9rc1
macOS
macOS 15.6
Context
1,024
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 87.0 | 19.3 | 15.8 GB | current |
| 4k | 89.0 | 18.6 | 17.3 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 19.3 | 1.00× |
| 2× | 22.1 | 1.15× |
| 4× | 43.9 | 2.27× |