# Can I run DeepSeek-V2-Lite on AMD Radeon RX 7900 XTX (24GB)?

Updated: 2026-06-15

**Yes, it runs.** Runs at Q4_K_M using ~12.2 GB of ~23 GB usable. You have room for Q8_0 for higher quality.

- Model: 16B (MoE, 2.4B active), Q4_K_M 10.4 GB
- Device: 24 GB vram, ~23 GB usable for weights
- Needs ~12.2 GB at Q4_K_M; recommended quant: Q4_K_M
- Best tool on Windows: LM Studio
- Command: `ollama run deepseek-v2:16b`

Estimate. Method: weights + KV cache + ~0.8GB overhead. Sources: https://ollama.com/library/deepseek-v2, https://huggingface.co/deepseek-ai/DeepSeek-V2-Lite, https://huggingface.co/mradermacher/DeepSeek-V2-Lite-GGUF, https://github.com/deepseek-ai/DeepSeek-V2.

More: https://localmodel.run/can-i-run/deepseek-v2-lite/amd-rx-7900-xtx-24gb