← Back to community benchmarks
Qwen3.5-27B-Claude-4.6-Opus-Distilled
Performance
8k
tokens
180.0
PP tok/s
28.8
TG tok/s
45515
TTFT (ms)
16.6
Peak mem (GB)
Hardware
Chip
M2 Ultra (60c)
Memory
64 GB
GPU Cores
60
Software
oMLX
v0.2.6
macOS
macOS 26.3
Context
8,192
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 173.5 | 32.7 | 15.8 GB | view |
| 4k | 180.8 | 30.8 | 16.1 GB | view |
| 8k | 180.0 | 28.8 | 16.6 GB | current |
| 16k | 176.9 | 25.6 | 17.5 GB | view |
| 32k | 169.8 | 20.6 | 19.2 GB | view |
| 64k | 154.9 | 11.2 | 22.8 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 32.7 | 1.00× |
| 2× | 41.5 | 1.27× |
| 4× | 46.2 | 1.41× |
| 8× | 48.8 | 1.49× |