← Back to community benchmarks
Qwen3.5-35B-A3B-Heretic
Performance
64k
tokens
586.3
PP tok/s
29.9
TG tok/s
111783
TTFT (ms)
21.9
Peak mem (GB)
Hardware
Chip
M1 Ultra (48c)
Memory
128 GB
GPU Cores
48
Software
oMLX
v0.2.6
macOS
macOS 26.3.1
Context
65,536
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 852.6 | 60.2 | 18.6 GB | view |
| 4k | 890.9 | 58.6 | 18.8 GB | view |
| 8k | 877.4 | 55.1 | 19.0 GB | view |
| 16k | 835.8 | 49.4 | 19.4 GB | view |
| 32k | 756.1 | 41.1 | 20.2 GB | view |
| 64k | 586.3 | 29.9 | 21.9 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 60.2 | 1.00× |
| 2× | 98.2 | 1.63× |
| 4× | 140.9 | 2.34× |
| 8× | 180.5 | 3.00× |