Open-source alternatives to Gemini vision

Gemini vision in the AI category. These are the open alternatives we recommend looking at.

Qwen2.5-VL-7B-Instruct

Vision-language model from Qwen that understands images, documents and video and answers in text.

Apache-2.0Permissive

SaaS: YesClosed product: Yes

Deployment
transformers

SmolVLM-Instruct

Small vision-language model from Hugging Face that describes and answers questions about images. Based on HuggingFaceTB/SmolLM2-1.7B-Instruct.

Apache-2.0Permissive

SaaS: YesClosed product: Yes

Deployment
transformers

This is guidance, not legal advice (inte juridisk rådgivning). Check with a lawyer before deciding.

Sign in for more

  • Full licence analysis per use case (self-hosting, modifying, hosted service, AI use)
  • Save products to your own lists
  • Systemkarta: map the tools you already use (coming soon)
  • Personal help and AI advice (coming soon)
Sign in for free