← Back to community benchmarks
Qwen3.5-35B-A3B
Performance
4k
tokens
1,125
PP tok/s
65.7
TG tok/s
3640
TTFT (ms)
36.3
Peak mem (GB)
Hardware
Chip
M2 Ultra (60c)
Memory
64 GB
GPU Cores
60
Software
oMLX
v0.2.6
macOS
macOS 15.2
Context
4,096
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 1,075 | 68.3 | 36.2 GB | view |
| 4k | 1,125 | 65.7 | 36.3 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 68.3 | 1.00× |
| 2× | 117.1 | 1.71× |
| 4× | 180.7 | 2.65× |
| 8× | 264.4 | 3.87× |