← Back to community benchmarks
Qwen3.5-35B-A3B-Claude-4.6-Opus-Reasoning-Distilled
Performance
8k
tokens
855.0
PP tok/s
74.3
TG tok/s
9582
TTFT (ms)
19.6
Peak mem (GB)
Hardware
Chip
M4 Pro (20c)
Memory
64 GB
GPU Cores
20
Software
oMLX
v0.2.12
macOS
macOS 26.3.1
Context
8,192
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 736.7 | 83.6 | 19.2 GB | view |
| 4k | 861.4 | 80.7 | 19.4 GB | view |
| 8k | 855.0 | 74.3 | 19.6 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 83.6 | 1.00× |
| 2× | 134.1 | 1.60× |
| 4× | 158.4 | 1.89× |
| 8× | 195.8 | 2.34× |