← Back to community benchmarks
Qwen3.6-27B-oQ6-fp16-mtp
Performance
128k
tokens
210.7
PP tok/s
18.5
TG tok/s
621993
TTFT (ms)
48.3
Peak mem (GB)
Hardware
Chip
M3 Ultra (60c)
Memory
256 GB
GPU Cores
60
Software
oMLX
v0.3.9.dev2
macOS
macOS 26.4.1
Context
131,072
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 319.6 | 33.1 | 23.0 GB | view |
| 4k | 339.3 | 32.4 | 24.4 GB | view |
| 8k | 337.5 | 31.1 | 25.4 GB | view |
| 16k | 328.1 | 29.2 | 26.9 GB | view |
| 32k | 308.1 | 27.5 | 29.9 GB | view |
| 64k | 268.9 | 22.9 | 36.0 GB | view |
| 128k | 210.7 | 18.5 | 48.3 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 33.1 | 1.00× |
| 2× | 39.7 | 1.20× |
| 4× | 43.9 | 1.33× |