QwenQwen3-30B-A3B on AMD Radeon RX 9060 XT 16GB
AMD Radeon RX 9060 XT 16GB runs Qwen3-30B-A3B at Q3_K_M due to limited memory bandwidth. Slow but functional — see all options below.
Model Size
30B
Device VRAM
16 GB
Bandwidth
320 GB/s
Quantization
Q3_K_M
Performance by Quantization
OwnRig currently has one published compatibility entry for Qwen3-30B-A3B on AMD Radeon RX 9060 XT 16GB at Q3_K_M. This is the best supported pairing we can stand behind today.
| Quantization | Speed | TTFT | Fits in VRAM | Rating | Confidence |
|---|---|---|---|---|---|
| Q3_K_M | – | – | ✓ Yes | Marginal | estimated |
Notes
Q3_K_M
RDNA 4 card with improving ROCm support; Vulkan remains the safer default in llama.cpp. Q3_K_M fits in VRAM on this card. MoE models still have to load the full expert pool into memory.
About Qwen3-30B-A3B
Qwen3-30B-A3B (30B) is a chat, coding, ai coding, reasoning, multi-purpose model. Mixture-of-Experts architecture with ~3B active parameters per token and ~30B total; inference still loads the full expert pool for typical local stacks, so VRAM tracks total model size while compute per token stays efficient. High efficiency for its quality tier. Apache 2.0; 32K default and 128K max context.
View all Qwen3-30B-A3B hardware options →About AMD Radeon RX 9060 XT 16GB
AMD Radeon RX 9060 XT 16GB has 16 GB at 320 GB/s. Street price: $349.
See all models AMD Radeon RX 9060 XT 16GB can run →Estimate method: Estimated from model VRAM requirements and AMD runtime maturity. Reference hardware source: github.com (2026-05-26)
Performance varies by driver version, inference engine, quantization method, context length, and system configuration. Figures shown are estimates based on community benchmarks and may not reflect your exact setup. Product names are trademarks of their respective owners. OwnRig is independent and not affiliated with any hardware or AI model provider.