← Back to community benchmarks
Qwen3.5-27B-mxfp8
Performance
1k
tokens
410.6
PP tok/s
10.1
TG tok/s
2494
TTFT (ms)
28.4
Peak mem (GB)
Hardware
Chip
M5 Pro (20c)
Memory
64 GB
GPU Cores
20
Software
oMLX
v0.2.10
macOS
macOS 26.3.1
Context
1,024
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 410.6 | 10.1 | 28.4 GB | current |
| 4k | 437.2 | 10.0 | 28.7 GB | view |
| 8k | 432.2 | 9.8 | 29.2 GB | view |
| 16k | 418.4 | 9.2 | 30.0 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 10.1 | 1.00× |
| 2× | 18.7 | 1.85× |
| 4× | 24.4 | 2.42× |