← Back to community benchmarks
Qwen3.5-27B-Claude-4.6-Opus-Distilled
Performance
1k
tokens
159.5
PP tok/s
17.5
TG tok/s
6421
TTFT (ms)
15.8
Peak mem (GB)
Hardware
Chip
M3 Max (30c)
Memory
36 GB
GPU Cores
30
Software
oMLX
v0.3.5.dev1
macOS
macOS 26.2
Context
1,024
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 159.5 | 17.5 | 15.8 GB | current |
| 4k | 161.2 | 16.8 | 17.3 GB | view |
| 32k | 139.5 | 12.1 | 21.6 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 17.5 | 1.00× |
| 2× | 26.0 | 1.49× |
| 4× | 29.6 | 1.69× |