Hardware
Radeon RX 9060 XT 16GB for local LLMs
16 GB new card. The stored price is the GPU PRIX board-list low, not the lower Amazon listing in the same feed.
Memory
16 GB VRAMBandwidth
320 GB/swww.amd.com/en/products/graphics/desktops/radeon/9000-series/amd-radeon-rx-9060xt.htmlPrice
$631€564
new (lowest in-stock DE board)
price as of 17 Oct 2026
GPU PRIXGPU PRIX Germany, last updated 17 Oct 2026 19:49 UTC. The board list's cheapest in-stock board is the ASUS Prime at €563.90. The offers feed also marks an Amazon.de XFX Swift in stock at €474.78, so the header near €475 is not only out-of-stock listings on this check. 3DCenter's October 2026 range is €520–590 (price range, since November 2025). Bandwidth is up to 320 GB/s on AMD's spec page. This catalog's AMD utilisation is 0.4, so decode is slower than the bandwidth suggests. ROCm/Vulkan, not CUDA.
Parts in the catalog
- ASUS Prime Radeon RX 9060 XT 16GB, the cheapest in-stock board on the GPU PRIX Germany board list
Which models fit
Estimate at 8,192 tokens, or the model's maximum when that is shorter. full means the formula fits in usable memory. offload means GPU plus system RAM. no means it does not fit. This is not a measured file.
| Model | Q4Q4_K_Mestimate | Q5Q5 (~0.71 bytes/param)estimate | Q8Q8estimate | FP16FP16estimate |
|---|---|---|---|---|
| Llama 3.2 3B | full | full | full | full |
| Llama 3.1 8B | full | full | full | no |
| Mistral 7B v0.3 | full | full | full | no |
| Gemma 2 9B | full | full | full | no |
| Qwen2.5 14B | full | full | no | no |
| Phi-4 14B | full | full | no | no |
| Gemma 2 27B | no | no | no | no |
| Qwen2.5 32B | no | no | no | no |
| Llama 3.3 70B | no | no | no | no |
| Qwen2.5 72B | no | no | no | no |
| DeepSeek-R1-Distill 70B | no | no | no | no |
| NVIDIA Nemotron 3.5 Bolt 74B-A9B | no | no | no | no |
| Reflection Beam 1.0 | no | no | no | no |
| NVIDIA Nemotron 3.5 Super VL | no | no | no | no |
| DeepSeek V4.1 Flash 1017 | no | no | no | no |
| Meta Muse Spark Open | no | no | no | no |
We'll match you with vetted providers
Embed this
Badges cover nine scored GPUs. This machine is not one of them.
Common questions
- Can Radeon RX 9060 XT 16GB run Llama 3.3 70B?
- No (estimate). Llama 3.3 70B at Q4_K_M needs about 46.0 GB at 8,192 tokens. Radeon RX 9060 XT 16GB has about 15.4 GB usable.
- Which models fit fully on Radeon RX 9060 XT 16GB at Q4?
- 6 of 16 models fit fully at Q4 (estimate, 8,192 tokens): Llama 3.2 3B, Llama 3.1 8B, Mistral 7B v0.3, Gemma 2 9B, Qwen2.5 14B, Phi-4 14B.
- How fast is Radeon RX 9060 XT 16GB (estimate)?
- About 44 tok/s for Llama 3.2 3B at Q4 (estimate, not a benchmark).