← Back to community benchmarks
Qwen3.5-27B-Claude-4.6-Opus-Reasoning-Distilled-qx64-hi
Performance
1k
tokens
180.1
PP tok/s
6.3
TG tok/s
5685
TTFT (ms)
19.9
Peak mem (GB)
Hardware
Chip
M5 (10c)
Memory
32 GB
GPU Cores
10
Software
oMLX
v0.2.20.dev3
macOS
macOS 26.3.1
Context
1,024
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 180.1 | 6.3 | 19.9 GB | current |
| 4k | 183.3 | 6.3 | 21.5 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 6.3 | 1.00× |
| 2× | 12.0 | 1.90× |
| 4× | 12.6 | 2.00× |
| 8× | 12.9 | 2.05× |