← Back to community benchmarks
Qwen3-Coder-30B-A3B-Instruct5bit
Performance
64k
tokens
190.1
PP tok/s
15.8
TG tok/s
344789
TTFT (ms)
26.3
Peak mem (GB)
Hardware
Chip
M4 Max (40c)
Memory
64 GB
GPU Cores
40
Software
oMLX
v0.2.7
macOS
macOS 26.3.1
Context
65,536
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 1,286 | 90.7 | 20.1 GB | view |
| 4k | 1,337 | 75.7 | 20.6 GB | view |
| 8k | 1,116 | 64.3 | 20.9 GB | view |
| 16k | 730.7 | 43.1 | 21.7 GB | view |
| 32k | 398.7 | 26.3 | 23.2 GB | view |
| 64k | 190.1 | 15.8 | 26.3 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 90.7 | 1.00× |
| 2× | 106.6 | 1.18× |
| 4× | 128.0 | 1.41× |
| 8× | 152.1 | 1.68× |