← Back to community benchmarks
Qwen3.5-35B-A3B
Performance
32k
tokens
1,164
PP tok/s
31.4
TG tok/s
28141
TTFT (ms)
37.8
Peak mem (GB)
Hardware
Chip
M2 Ultra (76c)
Memory
192 GB
GPU Cores
76
Software
oMLX
v0.2.6
macOS
macOS 15.0
Context
32,768
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 1,198 | 56.0 | 36.2 GB | view |
| 4k | 1,345 | 53.8 | 36.3 GB | view |
| 8k | 1,351 | 47.9 | 36.5 GB | view |
| 16k | 1,263 | 45.2 | 37.0 GB | view |
| 32k | 1,164 | 31.4 | 37.8 GB | current |
| 64k | 937.2 | 23.5 | 39.5 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 56.0 | 1.00× |
| 2× | 102.5 | 1.83× |
| 4× | 172.8 | 3.09× |
| 8× | 261.0 | 4.66× |