← Back to community benchmarks
Qwen3.5-27B-Claude-4.6-Opus-Distilled
Performance
4k
tokens
404.6
PP tok/s
33.4
TG tok/s
10123
TTFT (ms)
16.1
Peak mem (GB)
Hardware
Chip
M3 Ultra (80c)
Memory
512 GB
GPU Cores
80
Software
oMLX
v0.2.6
macOS
macOS 26.3
Context
4,096
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 373.9 | 34.7 | 15.8 GB | view |
| 4k | 404.6 | 33.4 | 16.1 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 34.7 | 1.00× |
| 2× | 62.6 | 1.80× |
| 4× | 87.3 | 2.52× |