← Back to community benchmarks
Qwen3.6-27B-oQ4-mtp
Performance
1k
tokens
207.1
PP tok/s
33.0
TG tok/s
4943
TTFT (ms)
16.7
Peak mem (GB)
Hardware
Chip
M3 Max (40c)
Memory
64 GB
GPU Cores
40
Software
oMLX
v0.3.9.dev2
macOS
macOS 26.5
Context
1,024
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 207.1 | 33.0 | 16.7 GB | current |
| 4k | 211.8 | 29.2 | 18.2 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 33.0 | 1.00× |
| 2× | 34.1 | 1.03× |
| 4× | 41.9 | 1.27× |