← Back to community benchmarks
Qwen3.5-35B-A3B
Performance
4k
tokens
1,281
PP tok/s
64.3
TG tok/s
3199
TTFT (ms)
36.3
Peak mem (GB)
Hardware
Chip
M2 Ultra (76c)
Memory
192 GB
GPU Cores
76
Software
oMLX
v0.2.7
macOS
macOS 26.3
Context
4,096
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 945.9 | 68.1 | 36.2 GB | view |
| 4k | 1,281 | 64.3 | 36.3 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 68.1 | 1.00× |
| 2× | 120.3 | 1.77× |
| 4× | 192.5 | 2.83× |