← Back to community benchmarks
Qwen3.5-27B
Performance
32k
tokens
126.1
PP tok/s
13.1
TG tok/s
259825
TTFT (ms)
34.2
Peak mem (GB)
Hardware
Chip
M1 Ultra (48c)
Memory
64 GB
GPU Cores
48
Software
oMLX
v0.3.5
macOS
macOS 26.5
Context
32,768
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 128.1 | 18.2 | 28.3 GB | view |
| 4k | 133.5 | 17.6 | 29.8 GB | view |
| 8k | 133.1 | 16.8 | 30.4 GB | view |
| 16k | 131.0 | 15.3 | 31.7 GB | view |
| 32k | 126.1 | 13.1 | 34.2 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 18.2 | 1.00× |
| 2× | 28.4 | 1.56× |
| 4× | 33.9 | 1.86× |
| 8× | 37.2 | 2.04× |