← Back to community benchmarks
Qwen3.6-27B-oQ4-mtp
Performance
1k
tokens
240.1
PP tok/s
38.3
TG tok/s
4265
TTFT (ms)
16.8
Peak mem (GB)
Hardware
Chip
M4 Max (40c)
Memory
128 GB
GPU Cores
40
Software
oMLX
v0.3.9.dev1
macOS
macOS 26.4.1
Context
1,024
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 240.1 | 38.3 | 16.8 GB | current |
| 4k | 206.9 | 32.7 | 18.2 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 38.3 | 1.00× |
| 2× | 37.7 | 0.98× |
| 4× | 42.9 | 1.12× |
| 8× | 47.0 | 1.23× |