← Back to community benchmarks
Qwen3.5-27B-Claude-4.6-Opus-Reasoning-Distilled
Performance
32k
tokens
80.2
PP tok/s
9.5
TG tok/s
408660
TTFT (ms)
19.2
Peak mem (GB)
Hardware
Chip
M1 Max (32c)
Memory
64 GB
GPU Cores
32
Software
oMLX
v0.2.7
macOS
macOS 15.7.3
Context
32,768
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 84.5 | 17.2 | 15.8 GB | view |
| 4k | 85.8 | 17.7 | 16.1 GB | view |
| 8k | 87.0 | 15.5 | 16.6 GB | view |
| 16k | 85.0 | 13.1 | 17.5 GB | view |
| 32k | 80.2 | 9.5 | 19.2 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 17.2 | 1.00× |
| 2× | 21.9 | 1.27× |
| 4× | 22.5 | 1.31× |
| 8× | 17.5 | 1.02× |