← Back to community benchmarks
Qwen3.5-27B-Claude-4.6-Opus-Distilled
Performance
4k
tokens
466.5
PP tok/s
17.7
TG tok/s
8780
TTFT (ms)
17.4
Peak mem (GB)
Hardware
Chip
M5 Pro (20c)
Memory
64 GB
GPU Cores
20
Software
oMLX
v0.2.21
macOS
macOS 26.4
Context
4,096
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 430.5 | 18.1 | 15.8 GB | view |
| 4k | 466.5 | 17.7 | 17.4 GB | current |
| 8k | 453.8 | 17.1 | 18.0 GB | view |
| 16k | 406.5 | 15.8 | 19.3 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 18.1 | 1.00× |
| 2× | 32.3 | 1.78× |
| 4× | 41.0 | 2.27× |
| 8× | 39.4 | 2.18× |