← Back to community benchmarks
qwen3.6-27b-oQ8-mtp
Performance
4k
tokens
318.5
PP tok/s
32.6
TG tok/s
12861
TTFT (ms)
30.3
Peak mem (GB)
Hardware
Chip
M3 Ultra (60c)
Memory
256 GB
GPU Cores
60
Software
oMLX
v0.3.9rc1
macOS
macOS 26.2
Context
4,096
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 298.2 | 31.6 | 28.8 GB | view |
| 4k | 318.5 | 32.6 | 30.3 GB | current |
| 8k | 317.5 | 30.5 | 31.3 GB | view |
| 16k | 309.3 | 28.8 | 32.8 GB | view |
| 32k | 291.4 | 26.8 | 35.8 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 31.6 | 1.00× |
| 2× | 39.2 | 1.24× |
| 4× | 50.1 | 1.59× |