InternVL2_5-78B
InternVL2_5-78B is an AI model developed by Shanghai AI Lab, SenseTime, Tsinghua University, Nanjing University, Fudan University, Chinese University of Hong Kong (CUHK) and Shanghai Jiao Tong University (China and Hong Kong), first published in December 2024. It works in the multimodal, language and vision domain, on tasks such as visual question answering, language modeling/generation, question answering and character recognition (ocr).
Epoch AI has no training-compute estimate for this model. The model has 78,400,000,000 parameters. It was trained on roughly 120B datapoints.
Access: Open weights (restricted use). Its weights are openly available. It is built on top of Qwen2.5 Instruct (72B),InternViT-6B. The reference paper has 1,468 citations. Epoch AI rates the confidence of this record as confident.
- Organization
- Shanghai AI Lab, SenseTime, Tsinghua University, Nanjing University, Fudan University, Chinese University of Hong Kong (CUHK), Shanghai Jiao Tong University
- Country of organization
- China, Hong Kong
- Domain
- Multimodal, Language, Vision
- Task
- Visual question answering, Language modeling/generation, Question answering, Character recognition (OCR)
- Compute estimation method
- Operation counting
- Parameters
- 78,400,000,000
- Dataset size
- 120B
- Model accessibility
- Open weights (restricted use)
- Open weights
- Yes
- Base model
- Qwen2.5 Instruct (72B), InternViT-6B
- Citations
- 1,468
- Epoch confidence
- Confident
| 01 | Video-MME | Overall (no subtitles) | 0.72 |