GGUF Quantized LLaVA 1.6 Vicuna 7B

Updated quants and projector from PR #5267

Provided files

Name	Quant method	Bits	Size	Use case
llava-v1.6-vicuna-7b.Q3_K_XS.gguf	Q3_K_XS	3	2.77 GB	very small, high quality loss
llava-v1.6-vicuna-7b.Q3_K_M.gguf	Q3_K_M	3	3.3 GB	very small, high quality loss
llava-v1.6-vicuna-7b.Q4_K_M.gguf	Q4_K_M	4	4.08 GB	medium, balanced quality - recommended
llava-v1.6-vicuna-7b.Q5_K_S.gguf	Q5_K_S	5	4.65 GB	large, low quality loss - recommended
llava-v1.6-vicuna-7b.Q5_K_M.gguf	Q5_K_M	5	4.78 GB	large, very low quality loss - recommended
llava-v1.6-vicuna-7b.Q6_K.gguf	Q6_K	6	5.53 GB	very large, extremely low quality loss
llava-v1.6-vicuna-7b.Q8_0.gguf	Q8_0	8	7.16 GB	very large, extremely low quality loss - not recommended

ORIGINAL LLaVA Model Card

Model details

Model type: LLaVA is an open-source chatbot trained by fine-tuning LLM on multimodal instruction-following data. It is an auto-regressive language model, based on the transformer architecture. Base LLM: lmsys/vicuna-7b-v1.5

Model date: LLaVA-v1.6-Vicuna-7B was trained in December 2023.

Paper or resources for more information: https://llava-vl.github.io/

License

Where to send questions or comments about the model: https://github.com/haotian-liu/LLaVA/issues

Intended use

Primary intended uses: The primary use of LLaVA is research on large multimodal models and chatbots.

Primary intended users: The primary intended users of the model are researchers and hobbyists in computer vision, natural language processing, machine learning, and artificial intelligence.

Training dataset

558K filtered image-text pairs from LAION/CC/SBU, captioned by BLIP.
158K GPT-generated multimodal instruction-following data.
500K academic-task-oriented VQA data mixture.
50K GPT-4V data mixture.
40K ShareGPT data.

Evaluation dataset

A collection of 12 benchmarks, including 5 academic VQA benchmarks and 7 recent benchmarks specifically proposed for instruction-following LMMs.