← Back to community benchmarks
Qwen3.5-27B-Claude-4.6-Opus-Distilled
Performance
4k
tokens
201.5
PP tok/s
21.8
TG tok/s
20332
TTFT (ms)
16.1
Peak mem (GB)
Hardware
Chip
M4 Max (32c)
Memory
36 GB
GPU Cores
32
Software
oMLX
v0.2.7
macOS
macOS 26.3
Context
4,096
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 194.9 | 22.7 | 15.8 GB | view |
| 4k | 201.5 | 21.8 | 16.1 GB | current |
| 8k | 199.4 | 20.9 | 16.6 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 22.7 | 1.00× |
| 2× | 44.3 | 1.95× |
| 4× | 51.2 | 2.26× |
| 8× | 53.1 | 2.34× |