← Back to community benchmarks
Qwen3___6-35B-A3B
Performance
8k
tokens
448.3
PP tok/s
39.2
TG tok/s
18273
TTFT (ms)
20.2
Peak mem (GB)
Hardware
Chip
M4 (10c)
Memory
32 GB
GPU Cores
10
Software
oMLX
v0.3.6
macOS
macOS 26.4.1
Context
8,192
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 419.2 | 47.0 | 19.2 GB | view |
| 4k | 461.6 | 44.4 | 19.9 GB | view |
| 8k | 448.3 | 39.2 | 20.2 GB | current |
| 16k | 411.8 | 21.2 | 20.9 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 47.0 | 1.00× |
| 2× | 71.1 | 1.51× |