AMD Radeon RX 9060 XT 16GB for local LLMs
Written by Jakub Rusinowski · Last updated
With 16 GB at 320 GB/s, the RX 9060 XT 16GB runs 80 catalogued models at Q4_K_M with 8K context. The largest that fits is OLMo 2 13B Instruct (~15.8 GB), and the top pick is EuroLLM 22B at 9.3–19 tok/s.
Affordable 16 GB RDNA 4 card. Lower bandwidth than the 9070-series, but the VRAM capacity fits all 13–14B models at Q4. Good ROCm support on Linux.
Models that run on the RX 9060 XT 16GB
Q4_K_M, 8K context, 16 GB usable. Ranked by quality and speed.
| Model | Memory · marker = 16 GB | VRAM | Speed |
|---|---|---|---|
| EuroLLM 22B EuroLLM | 14.1 GB | 9.3–19 tok/s | |
| GPT-OSS 20B GPT-OSS | 13.6 GB | 46–95 tok/s | |
| Cosmos 3 Nano Cosmos 3 | 10.5 GB | 21–43 tok/s | |
| StarCoder 2 15B StarCoder 2 | 10.8 GB | 13–28 tok/s | |
| Qwen 3 14B Qwen 3 | 11.1 GB | 13–27 tok/s | |
| Phi-4 (14B) Phi-4 Family | 10.9 GB | 13–28 tok/s | |
| Cogito v1 14B Cogito v1 | 10.9 GB | 13–28 tok/s | |
| Ministral 3 14B Ministral 3 | 9.3 GB | 14–28 tok/s | |
| DeepSeek R1 Distill Qwen 14B DeepSeek R1 | 10.6 GB | 14–29 tok/s | |
| Mellum2.1 12B Thinking Mellum2.1 | 9 GB | 60–125 tok/s | |
| Gemma 4 12B (Unified) Gemma 4 | 8.7 GB | 17–35 tok/s | |
| Bielik PL 11B v3.0 Instruct Bielik | 7.4 GB | 17–35 tok/s |
Buy it or rent the same memory
Buy the card, or rent a GPU with the same memory by the hour to try models first.
or compare on Vast.ai from $0.35/hr (typical low · varies)
As an Amazon Associate we earn from qualifying purchases. Cloud GPU links are referral links — we may earn a commission at no extra cost to you.
Speed vs other GPUs
Llama 3.1 8B, Q4_K_M. Estimated ranges. How this is calculated
Specifications
Specs last updated 2026-07-12.
- Memory
- 16 GB
- Memory bandwidth
- 320 GB/s
- Architecture
- RDNA 4 Navi 44
- Series
- RX 9000-series
- Board power
- 160 W
- Release year
- 2025
- Launch price
- $349
- Usable for models
- 16 GB
Similar GPUs
Frequently asked questions
Can the AMD Radeon RX 9060 XT 16GB run local LLMs?
Yes. With 16 GB (16 GB usable by a model) it runs 80 of the catalogued models at Q4_K_M with 8K context; the largest is OLMo 2 13B Instruct, needing about 15.8 GB.
How fast is the AMD Radeon RX 9060 XT 16GB for AI inference?
It is estimated to run Llama 3.1 8B at 24–49 tok/s at Q4_K_M. Llama 3.3 70B does not fit: it needs about 44 GB against 16 GB usable. These are modelled estimates from memory bandwidth, not measurements; the methodology page shows the formula.
What LLMs can I run on 16 GB?
Among the best that fit: EuroLLM 22B, GPT-OSS 20B, Cosmos 3 Nano, StarCoder 2 15B, Qwen 3 14B. The quickest start is Ollama: ollama run qwen3:14b.
Keep going
Can I run it on the RX 9060 XT 16GB?
More for this card
Tools