InternVL is an open-source multi-modal LLM family from OpenGVLab with chat capabilities, APIs, and multiple releases. It targets GPT-4o-like performance using open models and provides installation/docs links.
Collecting history — the radar snapshots this repo daily. The trend line appears after 3 days of data (1 so far).
What it is
InternVL Family: Closing the Gap to Commercial Multimodal Models with Open-Source Suites —— A Pioneering Open-Source Alternative to GPT-5
How it works
The repository aggregates multi-modal models and a chat interface (InternVL-Chat) spanning multiple versions (e.g., InternVL-Chat-V1-2, V1-5, etc.), with documentation and tutorials for installation, chat data formats, and a Local Chat Demo. Release notes indicate progress across model scales (1B to 78B) and optimization methods (MPO, VisualPRM).
Getting started
- Installation guide and requirements: Installation Guide | requirements.txt
- Local chat demo: Streamlit Demo
- InternVL-Chat API: InternVL2.5 API
- Quick Start section anchor: #quick-start-with-huggingface
Commands to replicate setup are provided in the README under Get Started:
- "Installation: 🌱 Installation Guide | 📄 requirements.txt"
- Other setup/docs links are listed in the Documents section.
Recent releases
Latest releases include:
- v1.5.0 InternVL-Chat-V1.5.0 (2024-05-08)
- v1.2.3 InternVL-Chat-V1.2.3 (2024-03-04)
- v1.2.2 InternVL-Chat-V1.2.2 (2024-02-21)
- v1.2 InternVL-Chat-V1.2 (2024-02-13)
- v1.1 InternVL-Chat-V1.1 (2024-02-13)
Traction
Stars: 10117
Behind the repo
Not provided in the FACTS block beyond general project scope and linked announcements.
Caveats
- License: MIT
- Created: 2023-11-22
- Last push: 2025-09-22
- Open issues: 313
- Language: Python
- Topics include gpt, gpt-4o, gpt-4v, vision-language, multimodal, etc.
Notes: The README excerpt contains multiple project milestones and blog links, but no additional installation commands beyond those referenced in the Get Started section.