← Back to community benchmarks
Qwen3.5-35B-A3B
Performance
64k
tokens
1,267
PP tok/s
25.1
TG tok/s
51738
TTFT (ms)
39.5
Peak mem (GB)
Hardware
Chip
M3 Ultra (60c)
Memory
96 GB
GPU Cores
60
Software
oMLX
v0.2.13
macOS
macOS 26.3.1
Context
65,536
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 2,024 | 77.6 | 36.2 GB | view |
| 4k | 2,120 | 74.5 | 36.3 GB | view |
| 8k | 2,056 | 66.3 | 36.5 GB | view |
| 16k | 1,899 | 62.3 | 37.0 GB | view |
| 32k | 1,648 | 46.1 | 37.8 GB | view |
| 64k | 1,267 | 25.1 | 39.5 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 77.6 | 1.00× |
| 2× | 130.3 | 1.68× |
| 4× | 201.3 | 2.59× |