← Back to community benchmarks
Qwen3.6-35B-A3B-Claude-4.7-Opus-Reasoning-Distilled-TextoQ6
Performance
16k
tokens
3,752
PP tok/s
97.0
TG tok/s
4367
TTFT (ms)
29.6
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
16,384
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 2,070 | 101.3 | 27.6 GB | view |
| 4k | 3,597 | 98.9 | 28.4 GB | view |
| 8k | 3,843 | 96.3 | 28.7 GB | view |
| 16k | 3,752 | 97.0 | 29.6 GB | current |
| 32k | 3,237 | 89.1 | 31.1 GB | view |
| 64k | 2,325 | 74.6 | 34.1 GB | view |
| 128k | 1,417 | 56.4 | 40.1 GB | view |
| 195k | 982.1 | 45.4 | 47.0 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 101.3 | 1.00× |
| 2× | 153.9 | 1.52× |
| 4× | 225.9 | 2.23× |
| 8× | 446.0 | 4.40× |