← Back to community benchmarks
Qwen3-Coder-30B-A3B-Instruct-4bit-dwq-v2
Performance
4k
tokens
1,253
PP tok/s
76.5
TG tok/s
3269
TTFT (ms)
17.1
Peak mem (GB)
Hardware
Chip
M4 Max (32c)
Memory
36 GB
GPU Cores
32
Software
oMLX
v0.3.4
macOS
macOS 26.3
Context
4,096
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 1,194 | 95.8 | 16.6 GB | view |
| 4k | 1,253 | 76.5 | 17.1 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 95.8 | 1.00× |
| 2× | 139.6 | 1.46× |
| 4× | 185.3 | 1.93× |