← Back to community benchmarks
Qwen3.5-35B-A3B
Performance
16k
tokens
439.0
PP tok/s
41.5
TG tok/s
37325
TTFT (ms)
37.1
Peak mem (GB)
Hardware
Chip
M1 Max (24c)
Memory
64 GB
GPU Cores
24
Software
oMLX
v0.3.4
macOS
macOS 26.4
Context
16,384
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 396.7 | 52.6 | 35.4 GB | view |
| 4k | 459.1 | 49.8 | 36.2 GB | view |
| 8k | 457.4 | 46.6 | 36.5 GB | view |
| 16k | 439.0 | 41.5 | 37.1 GB | current |
| 32k | 399.7 | 34.4 | 38.5 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 52.6 | 1.00× |
| 2× | 76.4 | 1.45× |
| 4× | 95.1 | 1.81× |