← Back to community benchmarks
Qwen3.5-4B-Claude-4.6-Opus-Reasoning-Distilled
Performance
4k
tokens
248.3
PP tok/s
29.1
TG tok/s
16497
TTFT (ms)
3.5
Peak mem (GB)
Hardware
Chip
M4 (10c)
Memory
16 GB
GPU Cores
10
Software
oMLX
v0.2.13
macOS
macOS 26.2
Context
4,096
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 241.0 | 31.0 | 3.3 GB | view |
| 4k | 248.3 | 29.1 | 3.5 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 31.0 | 1.00× |
| 2× | 45.7 | 1.47× |
| 4× | 54.8 | 1.77× |