← Back to community benchmarks
Qwen3.6-27B-MTPLX-Optimized-Speed
Performance
1k
tokens
86.4
PP tok/s
19.1
TG tok/s
11847
TTFT (ms)
15.8
Peak mem (GB)
Hardware
Chip
M1 Max (32c)
Memory
64 GB
GPU Cores
32
Software
oMLX
v0.3.9.dev2
macOS
macOS 15.7.5
Context
1,024
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 86.4 | 19.1 | 15.8 GB | current |
| 4k | 88.3 | 18.3 | 17.3 GB | view |
| 8k | 87.6 | 17.0 | 18.3 GB | view |
| 16k | 75.6 | 13.1 | 19.8 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 19.1 | 1.00× |
| 2× | 17.4 | 0.91× |
| 4× | 18.3 | 0.96× |