← Back to community benchmarks
Qwen3.5-27B
Performance
4k
tokens
317.1
PP tok/s
2.6
TG tok/s
12918
TTFT (ms)
58.1
Peak mem (GB)
Hardware
Chip
M3 Ultra (80c)
Memory
512 GB
GPU Cores
80
Software
oMLX
v0.2.6
macOS
macOS 15.3
Context
4,096
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 247.7 | 2.6 | 57.7 GB | view |
| 4k | 317.1 | 2.6 | 58.1 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 2.6 | 1.00× |
| 2× | 4.4 | 1.69× |
| 4× | 8.7 | 3.35× |