← Back to community benchmarks
Qwen3.5-27B-Claude-4.6-Opus-Distilled
Performance
4k
tokens
190.0
PP tok/s
15.8
TG tok/s
21558
TTFT (ms)
22.4
Peak mem (GB)
Hardware
Chip
M3 Max (40c)
Memory
128 GB
GPU Cores
40
Software
oMLX
v0.2.8
macOS
macOS 26.3.1
Context
4,096
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 200.1 | 16.4 | 22.0 GB | view |
| 4k | 190.0 | 15.8 | 22.4 GB | current |
| 8k | 176.6 | 10.6 | 22.8 GB | view |
| 16k | 131.1 | 7.9 | 23.7 GB | view |
| 32k | 132.0 | 8.1 | 25.5 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 16.4 | 1.00× |
| 2× | 15.6 | 0.95× |
| 4× | 18.2 | 1.11× |
| 8× | 20.7 | 1.26× |