← Back to community benchmarks
Qwen3.6-35B-A3B
Performance
1k
tokens
1,140
PP tok/s
78.3
TG tok/s
898
TTFT (ms)
27.3
Peak mem (GB)
Hardware
Chip
M3 Max (40c)
Memory
128 GB
GPU Cores
40
Software
oMLX
v0.3.6
macOS
macOS 26.4.1
Context
1,024
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 1,140 | 78.3 | 27.3 GB | current |
| 4k | 1,430 | 75.8 | 28.1 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 78.3 | 1.00× |
| 2× | 124.1 | 1.58× |
| 4× | 168.4 | 2.15× |
| 8× | 171.2 | 2.19× |