# Best local LLMs for Nvidia GeForce RTX 5080 (16GB)

Updated: 2026-11-01

Nvidia GeForce RTX 5080 (16GB): 16 GB vram, ~15 GB usable. Runs 83 of 161 tracked models.

- gpt-oss 20B (21B): runs
- ERNIE 4.5 21B-A3B (21B): tight
- DeepSeek-V2-Lite (16B): runs
- Phi-4 14B (14B): runs
- Qwen2.5 14B (14B): runs
- Qwen3 14B (14B): runs
- DeepSeek-R1-Distill-Qwen 14B (14B): runs
- Qwen2.5 Coder 14B (14B): runs
- Phi-4-reasoning (14B): runs
- Ministral 3 14B (14B): runs
- Mistral Nemo 12B (12.2B): runs
- Gemma 3 12B (12B): runs

More: https://localmodel.run/best-llm-for/nvidia-rtx-5080-16gb