← Back to community benchmarks
Qwen3.5-35B-A3B-Claude-4.6-Opus-Reasoning-Distilled
Performance
4k
tokens
1,514
PP tok/s
81.0
TG tok/s
2706
TTFT (ms)
27.4
Peak mem (GB)
Hardware
Chip
M4 Max (40c)
Memory
128 GB
GPU Cores
40
Software
oMLX
v0.2.7
macOS
macOS 26.3.1
Context
4,096
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 1,168 | 83.5 | 27.3 GB | view |
| 4k | 1,514 | 81.0 | 27.4 GB | current |
| 8k | 1,493 | 79.9 | 27.6 GB | view |
| 16k | 1,432 | 62.8 | 28.1 GB | view |
| 32k | 1,248 | 43.3 | 28.9 GB | view |
| 64k | 961.0 | 24.1 | 30.6 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 83.5 | 1.00× |
| 2× | 137.3 | 1.64× |
| 4× | 193.2 | 2.31× |
| 8× | 238.7 | 2.86× |