← Back to community benchmarks
Qwen3.6-27B-oQ4-fp16-mtp
Performance
16k
tokens
217.3
PP tok/s
35.9
TG tok/s
75383
TTFT (ms)
20.4
Peak mem (GB)
Hardware
Chip
M4 Max (40c)
Memory
128 GB
GPU Cores
40
Software
oMLX
v0.3.9.dev2
macOS
macOS 26.4.1
Context
16,384
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 255.5 | 34.4 | 16.7 GB | view |
| 4k | 246.1 | 39.7 | 18.2 GB | view |
| 16k | 217.3 | 35.9 | 20.4 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 34.4 | 1.00× |
| 2× | 48.3 | 1.40× |
| 4× | 54.1 | 1.57× |