← Back to community benchmarks
Qwen3.6-27B-oQ8-mtp
Performance
4k
tokens
164.2
PP tok/s
4.5
TG tok/s
24951
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 26.3
Context
4,096
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 134.3 | 4.6 | 28.8 GB | view |
| 4k | 164.2 | 4.5 | 30.3 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 4.6 | 1.00× |
| 2× | 5.6 | 1.22× |
| 4× | 7.5 | 1.63× |