← Back to community benchmarks
Qwen3.5-27B
Performance
16k
tokens
70.0
PP tok/s
10.5
TG tok/s
233962
TTFT (ms)
18.3
Peak mem (GB)
Hardware
Chip
M1 Max (32c)
Memory
64 GB
GPU Cores
32
Software
oMLX
v0.2.13
macOS
macOS 15.7.4
Context
16,384
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 79.2 | 12.5 | 16.6 GB | view |
| 4k | 71.6 | 12.3 | 17.0 GB | view |
| 8k | 71.2 | 11.7 | 17.4 GB | view |
| 16k | 70.0 | 10.5 | 18.3 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 12.5 | 1.00× |
| 2× | 13.8 | 1.10× |
| 4× | 18.3 | 1.46× |