← Back to community benchmarks
Qwen3.5-27B
Performance
1k
tokens
382.9
PP tok/s
12.6
TG tok/s
2675
TTFT (ms)
51.7
Peak mem (GB)
Hardware
Chip
M3 Ultra (80c)
Memory
512 GB
GPU Cores
80
Software
oMLX
v0.3.6
macOS
macOS 26.4.1
Context
1,024
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 382.9 | 12.6 | 51.7 GB | current |
| 4k | 431.9 | 12.4 | 53.3 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 12.6 | 1.00× |
| 2× | 14.0 | 1.11× |
| 4× | 25.4 | 2.02× |