← Back to community benchmarks
Qwen3.5-27B-Claude-4.6-Opus-Distilled
Performance
8k
tokens
89.6
PP tok/s
16.2
TG tok/s
91408
TTFT (ms)
16.6
Peak mem (GB)
Hardware
Chip
M1 Max (32c)
Memory
64 GB
GPU Cores
32
Software
oMLX
v0.2.7
macOS
macOS 26.3.1
Context
8,192
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 88.6 | 19.2 | 15.8 GB | view |
| 4k | 90.0 | 17.0 | 16.1 GB | view |
| 8k | 89.6 | 16.2 | 16.6 GB | current |
| 16k | 88.0 | 9.7 | 17.5 GB | view |
| 32k | 84.3 | 6.4 | 19.2 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 19.2 | 1.00× |
| 2× | 22.5 | 1.17× |
| 4× | 17.5 | 0.91× |