← Back to community benchmarks
Qwen3.5-9B-Claude-4.6-Opus-Reasoning-Distilled
Performance
1k
tokens
215.6
PP tok/s
22.0
TG tok/s
4750
TTFT (ms)
5.7
Peak mem (GB)
Hardware
Chip
M4 (10c)
Memory
24 GB
GPU Cores
10
Software
oMLX
v0.2.13
macOS
macOS 26.3.1
Context
1,024
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 215.6 | 22.0 | 5.7 GB | current |
| 4k | 223.4 | 21.3 | 5.9 GB | view |
| 8k | 220.8 | 20.4 | 6.2 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 22.0 | 1.00× |
| 2× | 43.3 | 1.97× |
| 4× | 52.5 | 2.39× |