← Back to community benchmarks
Huihui-Qwen3.5-35B-A3B-abliterated
Performance
4k
tokens
2,016
PP tok/s
80.7
TG tok/s
2032
TTFT (ms)
19.5
Peak mem (GB)
Hardware
Chip
M3 Ultra (80c)
Memory
512 GB
GPU Cores
80
Software
oMLX
v0.2.6
macOS
macOS 26.3
Context
4,096
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 1,540 | 83.5 | 19.2 GB | view |
| 4k | 2,016 | 80.7 | 19.5 GB | current |
| 8k | 2,069 | 73.4 | 20.0 GB | view |
| 16k | 1,961 | 64.3 | 20.8 GB | view |
| 32k | 1,650 | 53.0 | 22.4 GB | view |
| 64k | 1,136 | 38.1 | 25.7 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 83.5 | 1.00× |
| 2× | 138.3 | 1.66× |
| 4× | 239.0 | 2.86× |
| 8× | 361.7 | 4.33× |