← Back to community benchmarks
Qwen3.6-27B-oQ8-mtp
Performance
4k
tokens
420.6
PP tok/s
31.8
TG tok/s
9739
TTFT (ms)
30.3
Peak mem (GB)
Hardware
Chip
M3 Ultra (80c)
Memory
512 GB
GPU Cores
80
Software
oMLX
v0.3.9.dev1
macOS
macOS 26.4.1
Context
4,096
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 386.8 | 32.1 | 28.8 GB | view |
| 4k | 420.6 | 31.8 | 30.3 GB | current |
| 8k | 420.1 | 31.0 | 31.3 GB | view |
| 16k | 409.4 | 29.0 | 32.8 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 32.1 | 1.00× |
| 2× | 38.6 | 1.20× |
| 4× | 57.7 | 1.80× |