← Back to community benchmarks
Qwen3.5-35B-A3B
Performance
16k
tokens
548.4
PP tok/s
27.2
TG tok/s
29875
TTFT (ms)
37.0
Peak mem (GB)
Hardware
Chip
M1 Max (32c)
Memory
64 GB
GPU Cores
32
Software
oMLX
v0.2.7
macOS
macOS 26.3.1
Context
16,384
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 549.4 | 50.2 | 36.2 GB | view |
| 4k | 583.2 | 47.9 | 36.3 GB | view |
| 8k | 573.1 | 45.0 | 36.5 GB | view |
| 16k | 548.4 | 27.2 | 37.0 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 50.2 | 1.00× |
| 2× | 79.4 | 1.58× |
| 4× | 111.2 | 2.22× |