← Back to community benchmarks
Qwen3.5-35B-A3B9bit
Performance
32k
tokens
422.7
PP tok/s
30.4
TG tok/s
77512
TTFT (ms)
40.7
Peak mem (GB)
Hardware
Chip
M1 Max (32c)
Memory
64 GB
GPU Cores
32
Software
oMLX
v0.2.18
macOS
macOS 26.3.1
Context
32,768
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 501.8 | 51.8 | 37.4 GB | view |
| 4k | 551.2 | 49.5 | 38.4 GB | view |
| 8k | 513.0 | 46.9 | 38.7 GB | view |
| 16k | 477.3 | 39.4 | 39.3 GB | view |
| 32k | 422.7 | 30.4 | 40.7 GB | current |
| 64k | 343.8 | 21.3 | 43.3 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 51.8 | 1.00× |
| 2× | 78.4 | 1.51× |
| 4× | 99.2 | 1.92× |
| 8× | 103.7 | 2.00× |