← Back to community benchmarks
Qwen3.6-27B-oQ4-fp16-mtp
Performance
1k
tokens
62.6
PP tok/s
5.8
TG tok/s
16359
TTFT (ms)
16.4
Peak mem (GB)
Hardware
Chip
M4 (10c)
Memory
32 GB
GPU Cores
10
Software
oMLX
v0.3.9rc1
macOS
macOS 26.5
Context
1,024
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 62.6 | 5.8 | 16.4 GB | current |
| 4k | 58.9 | 6.0 | 17.8 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 5.8 | 1.00× |
| 2× | 12.1 | 2.09× |