← Back to community benchmarks
Qwen3-Coder-REAP-25B-A3B-4bit
Performance
1k
tokens
404.3
PP tok/s
41.3
TG tok/s
2533
TTFT (ms)
13.6
Peak mem (GB)
Hardware
Chip
M4 (10c)
Memory
32 GB
GPU Cores
10
Software
oMLX
v0.2.6
macOS
macOS 26.3.1
Context
1,024
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 404.3 | 41.3 | 13.6 GB | current |
| 4k | 355.2 | 32.3 | 14.0 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 41.3 | 1.00× |
| 2× | 54.3 | 1.31× |
| 4× | 63.0 | 1.53× |