self-hosted/ai
§01·spec · /gpus

RTX 3080 Ti

nvidia30 series12GB VRAM

31 open-weights AI models run on the RTX 3080 Ti — see which fit, how fast they go, and the VRAM each needs.

§02·models that run on this GPU
31 total
LLM · 5
ModelBest speedMin VRAMWorksEvidence
Qwen3-8B115.2tokens/s12GB2benchesrecipecheck ↗
Qwen3 14B69.9tokens/s12GB2benchesrecipecheck ↗
gpt-oss 20B12GBrecipecheck ↗
Llama 3.1 8B10GBrecipecheck ↗
Ornith 1.0 9B12GBrecipecheck ↗
Multimodal · 2
ModelBest speedMin VRAMWorksEvidence
Gemma 4 E4B-IT6GBrecipecheck ↗
MiniMind-O4GBrecipecheck ↗
Video · 2
ModelBest speedMin VRAMWorksEvidence
LightX2V12GBrecipecheck ↗
WAN 2.28GBrecipecheck ↗
3D · 1
ModelBest speedMin VRAMWorksEvidence
Hunyuan3D10GBrecipecheck ↗
specialized · 2
ModelBest speedMin VRAMWorksEvidence
KiMoDo3GBrecipecheck ↗
SAM 34GBrecipecheck ↗
§03·tested recipes
showing 6 of 31