← Back to community benchmarks
Qwen3.6-35B-A3B-Claude-4.6-Opus-Reasoning-Distilled
Performance
1k
tokens
2,056
PP tok/s
127.2
TG tok/s
498
TTFT (ms)
19.3
Peak mem (GB)
Hardware
Chip
M5 Max (40c)
Memory
128 GB
GPU Cores
40
Software
oMLX
v0.3.8
macOS
macOS 26.4.1
Context
1,024
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 2,056 | 127.2 | 19.3 GB | current |
| 4k | 3,583 | 118.3 | 20.0 GB | view |
| 8k | 3,949 | 114.2 | 20.4 GB | view |
| 16k | 3,725 | 104.0 | 21.2 GB | view |
| 32k | 3,178 | 102.3 | 22.7 GB | view |
| 64k | 2,278 | 82.0 | 25.7 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 127.2 | 1.00× |
| 2× | 203.6 | 1.60× |
| 4× | 316.2 | 2.49× |
| 8× | 419.7 | 3.30× |