← Back to community benchmarks
Qwen3.5-27B-Claude-4.6-Opus-Distilled
Performance
4k
tokens
60.8
PP tok/s
6.6
TG tok/s
67370
TTFT (ms)
16.1
Peak mem (GB)
Hardware
Chip
M4 (10c)
Memory
32 GB
GPU Cores
10
Software
oMLX
v0.2.7
macOS
macOS 26.2
Context
4,096
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 62.7 | 6.9 | 15.7 GB | view |
| 4k | 60.8 | 6.6 | 16.1 GB | current |
| 8k | 58.5 | 6.5 | 16.5 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 6.9 | 1.00× |
| 2× | 13.4 | 1.94× |
| 4× | 15.6 | 2.26× |
| 8× | 16.2 | 2.35× |