← Back to community benchmarks
Qwen3.5-4B-Claude-4.6-Opus-Reasoning-Distilled
Performance
4k
tokens
389.3
PP tok/s
8.9
TG tok/s
10521
TTFT (ms)
6.2
Peak mem (GB)
Hardware
Chip
M4 (10c)
Memory
16 GB
GPU Cores
10
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 | 366.3 | 20.9 | 5.1 GB | view |
| 4k | 389.3 | 8.9 | 6.2 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 20.9 | 1.00× |
| 2× | 37.9 | 1.81× |
| 4× | 68.5 | 3.28× |