← Back to community benchmarks
Qwen3.6-27B-oQ4-mtp
Performance
16k
tokens
688.3
PP tok/s
37.3
TG tok/s
23802
TTFT (ms)
20.7
Peak mem (GB)
Hardware
Chip
M5 Max (40c)
Memory
64 GB
GPU Cores
40
Software
oMLX
v0.3.9.dev2
macOS
macOS 26.4.1
Context
16,384
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 790.2 | 47.6 | 16.7 GB | view |
| 4k | 810.1 | 44.7 | 18.2 GB | view |
| 8k | 723.8 | 40.5 | 19.2 GB | view |
| 16k | 688.3 | 37.3 | 20.7 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 47.6 | 1.00× |
| 2× | 49.5 | 1.04× |
| 4× | 60.6 | 1.27× |