Viscon Contextual Captioner is an open-source language model by tiiuae. Features: 8.4b LLM, VRAM: 33.5GB, LLM Explorer Score: 0.13.
| LLM Name | Viscon Contextual Captioner |
| Repository 🤗 | https://huggingface.co/tiiuae/viscon-contextual-captioner |
| Model Size | 8.4b |
| Required VRAM | 33.5 GB |
| Updated | 2026-08-08 |
| Maintainer | tiiuae |
| Model Type | idefics2 |
| Model Files | |
| Model Architecture | Idefics2ForConditionalGeneration |
| Transformers Version | 4.41.0 |
| Tokenizer Class | LlamaTokenizer |
| Padding Token | <unk> |
| Torch Data Type | float32 |