← Back to community benchmarks
Qwen3.5-27B
Performance
32k
tokens
75.3
PP tok/s
9.0
TG tok/s
435281
TTFT (ms)
20.1
Peak mem (GB)
Hardware
Chip
M2 Max (30c)
Memory
64 GB
GPU Cores
30
Software
oMLX
v0.2.10
macOS
macOS 26.2
Context
32,768
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 91.7 | 14.8 | 16.6 GB | view |
| 4k | 85.7 | 13.2 | 17.0 GB | view |
| 8k | 81.7 | 12.3 | 17.4 GB | view |
| 16k | 79.4 | 11.1 | 18.3 GB | view |
| 32k | 75.3 | 9.0 | 20.1 GB | current |
| 64k | 70.9 | 6.3 | 23.6 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 14.8 | 1.00× |
| 2× | 16.7 | 1.13× |
| 4× | 19.4 | 1.31× |
| 8× | 16.5 | 1.11× |