← Back to community benchmarks
Qwen3.5-27B-Claude-4.6-Opus-Distilled
Performance
4k
tokens
126.5
PP tok/s
6.4
TG tok/s
32390
TTFT (ms)
17.2
Peak mem (GB)
Hardware
Chip
M5 (10c)
Memory
24 GB
GPU Cores
10
Software
oMLX
v0.3.6
macOS
macOS 26.3.1
Context
4,096
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 122.9 | 6.6 | 15.8 GB | view |
| 4k | 126.5 | 6.4 | 17.2 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 6.6 | 1.00× |
| 2× | 10.4 | 1.58× |
| 4× | 12.2 | 1.85× |