← Back to community benchmarks
Qwen3.5-35B-A3Blm
Performance
8k
tokens
1,128
PP tok/s
70.2
TG tok/s
7262
TTFT (ms)
18.6
Peak mem (GB)
Hardware
Chip
M2 Ultra (60c)
Memory
192 GB
GPU Cores
60
Software
oMLX
v0.2.10
macOS
macOS 26.1
Context
8,192
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 681.5 | 77.7 | 18.2 GB | view |
| 4k | 1,049 | 74.4 | 18.4 GB | view |
| 8k | 1,128 | 70.2 | 18.6 GB | current |
| 16k | 1,092 | 60.4 | 19.0 GB | view |
| 32k | 993.4 | 50.2 | 19.9 GB | view |
| 64k | 820.3 | 27.8 | 21.5 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 77.7 | 1.00× |
| 2× | 128.9 | 1.66× |
| 4× | 186.8 | 2.40× |
| 8× | 235.6 | 3.03× |