← Back to community benchmarks
Huihui-Qwen3.5-27B-abliterated-fp16
Performance
4k
tokens
234.8
PP tok/s
8.4
TG tok/s
17447
TTFT (ms)
52.8
Peak mem (GB)
Hardware
Chip
M4 Max (40c)
Memory
128 GB
GPU Cores
40
Software
oMLX
v0.2.13
macOS
macOS 15.7
Context
4,096
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 262.2 | 9.1 | 52.5 GB | view |
| 4k | 234.8 | 8.4 | 52.8 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 9.1 | 1.00× |
| 2× | 5.8 | 0.64× |
| 4× | 10.7 | 1.18× |
| 8× | 20.7 | 2.27× |