← Back to community benchmarks
Qwen3.5-27B-Claude-4.6-Opus-Distilled
Performance
32k
tokens
169.9
PP tok/s
20.7
TG tok/s
192873
TTFT (ms)
19.2
Peak mem (GB)
Hardware
Chip
M2 Ultra (60c)
Memory
64 GB
GPU Cores
60
Software
oMLX
v0.2.10
macOS
macOS 26.3.1
Context
32,768
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 174.6 | 31.9 | 15.8 GB | view |
| 4k | 180.8 | 30.2 | 16.1 GB | view |
| 8k | 180.2 | 28.7 | 16.6 GB | view |
| 16k | 177.0 | 25.9 | 17.5 GB | view |
| 32k | 169.9 | 20.7 | 19.2 GB | current |
| 64k | 155.0 | 10.4 | 22.8 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 31.9 | 1.00× |
| 2× | 40.6 | 1.27× |
| 4× | 45.2 | 1.42× |