← Back to community benchmarks
Qwen3.6-27B-oQ4-mtp
Performance
1k
tokens
194.1
PP tok/s
30.8
TG tok/s
5275
TTFT (ms)
16.8
Peak mem (GB)
Hardware
Chip
M4 Max (32c)
Memory
36 GB
GPU Cores
32
Software
oMLX
v0.3.9.dev1
macOS
macOS 26.4
Context
1,024
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 194.1 | 30.8 | 16.8 GB | current |
| 4k | 166.0 | 26.3 | 18.2 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 30.8 | 1.00× |
| 2× | 30.7 | 1.00× |
| 4× | 36.4 | 1.18× |
| 8× | 30.6 | 0.99× |