← Back to community benchmarks
Qwen3-Coder-30B-A3B-Instruct
Performance
4k
tokens
1,022
PP tok/s
46.4
TG tok/s
4006
TTFT (ms)
17.0
Peak mem (GB)
Hardware
Chip
M5 (10c)
Memory
32 GB
GPU Cores
10
Software
oMLX
v0.2.12
macOS
macOS 26.3.1
Context
4,096
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 986.2 | 55.1 | 16.6 GB | view |
| 4k | 1,022 | 46.4 | 17.0 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 55.1 | 1.00× |
| 2× | 71.8 | 1.30× |
| 4× | 81.3 | 1.48× |