--- base_model: llava-hf/llava-v1.6-mistral-7b-hf language: - en library_name: transformers pipeline_tag: image-text-to-text license: apache-2.0 tags: - multimodal - llava - vision - unsloth - mistral --- # Finetune Llama 3.2, Qwen 2.5, Gemma 2, Mistral 2-5x faster with 70% less memory via Unsloth! We have a free Google Colab Tesla T4 notebook for Llava 1.6 (7B) here: https://colab.research.google.com/drive/1Ys44kVvmeZtnICzWz0xgpRnrIOjZAuxp?usp=sharing And a free notebook for [Llama 3.2 Vision (11B) here](https://colab.research.google.com/drive/1Ys44kVvmeZtnICzWz0xgpRnrIOjZAuxp?usp=sharing) [](https://discord.gg/unsloth) [](https://github.com/unslothai/unsloth) # unsloth/llava-v1.6-mistral-7b-hf-bnb-4bit For more details on the model, please go to the original [model card](https://huggingface.co/llava-hf/llava-v1.6-mistral-7b-hf) ## ✨ Finetune for Free All notebooks are **beginner friendly**! Add your dataset, click "Run All", and you'll get a 2x faster finetuned model which can be exported to GGUF, vLLM or uploaded to Hugging Face. | Unsloth supports | Free Notebooks | Performance | Memory use | |-----------------|--------------------------------------------------------------------------------------------------------------------------|-------------|----------| | **Llama-3.2 (3B)** | [▶️ Start on Colab](https://colab.research.google.com/drive/1Ys44kVvmeZtnICzWz0xgpRnrIOjZAuxp?usp=sharing) | 2.4x faster | 58% less | | **Llama-3.2 (11B vision)** | [▶️ Start on Colab](https://colab.research.google.com/drive/1Ys44kVvmeZtnICzWz0xgpRnrIOjZAuxp?usp=sharing) | 2x faster | 40% less | | **Qwen2 VL (7B)** | [▶️ Start on Colab](https://colab.research.google.com/drive/1Ys44kVvmeZtnICzWz0xgpRnrIOjZAuxp?usp=sharing) | 1.8x faster | 40% less | | **Qwen2.5 (7B)** | [▶️ Start on Colab](https://colab.research.google.com/drive/1Kose-ucXO1IBaZq5BvbwWieuubP7hxvQ?usp=sharing) | 2x faster | 60% less | | **Llama-3.1 (8B)** | [▶️ Start on Colab](https://colab.research.google.com/drive/1Ys44kVvmeZtnICzWz0xgpRnrIOjZAuxp?usp=sharing) | 2.4x faster | 58% less | | **Phi-3.5 (mini)** | [▶️ Start on Colab](https://colab.research.google.com/drive/1lN6hPQveB_mHSnTOYifygFcrO8C1bxq4?usp=sharing) | 2x faster | 50% less | | **Gemma 2 (9B)** | [▶️ Start on Colab](https://colab.research.google.com/drive/1vIrqH5uYDQwsJ4-OO3DErvuv4pBgVwk4?usp=sharing) | 2.4x faster | 58% less | | **Mistral (7B)** | [▶️ Start on Colab](https://colab.research.google.com/drive/1Dyauq4kTZoLewQ1cApceUQVNcnnNTzg_?usp=sharing) | 2.2x faster | 62% less | | **DPO - Zephyr** | [▶️ Start on Colab](https://colab.research.google.com/drive/15vttTpzzVXv_tJwEk-hIcQ0S9FcEWvwP?usp=sharing) | 1.9x faster | 19% less | [](https://docs.unsloth.ai) - This [conversational notebook](https://colab.research.google.com/drive/1Aau3lgPzeZKQ-98h69CCu1UJcvIBLmy2?usp=sharing) is useful for ShareGPT ChatML / Vicuna templates. - This [text completion notebook](https://colab.research.google.com/drive/1ef-tab5bhkvWmBOObepl1WgJvfvSzn5Q?usp=sharing) is for raw text. This [DPO notebook](https://colab.research.google.com/drive/15vttTpzzVXv_tJwEk-hIcQ0S9FcEWvwP?usp=sharing) replicates Zephyr. - \* Kaggle has 2x T4s, but we use 1. Due to overhead, 1x T4 is 5x faster. # LLaVa-Next, leveraging [mistralai/Mistral-7B-Instruct-v0.2](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.2) as LLM The LLaVA-NeXT model was proposed in [LLaVA-NeXT: Improved reasoning, OCR, and world knowledge](https://llava-vl.github.io/blog/2024-01-30-llava-next/) by Haotian Liu, Chunyuan Li, Yuheng Li, Bo Li, Yuanhan Zhang, Sheng Shen, Yong Jae Lee. LLaVa-NeXT (also called LLaVa-1.6) improves upon [LLaVa-1.5](https://huggingface.co/transformers/main/model_doc/llava.html) by increasing the input image resolution and training on an improved visual instruction tuning dataset to improve OCR and common sense reasoning. Disclaimer: The team releasing LLaVa-NeXT did not write a model card for this model so this model card has been written by the Hugging Face team. ## Model description LLaVa combines a pre-trained large language model with a pre-trained vision encoder for multimodal chatbot use cases. LLaVA 1.6 improves on LLaVA 1.5 BY: - Using [Mistral-7B](https://mistral.ai/news/announcing-mistral-7b/) (for this checkpoint) and [Nous-Hermes-2-Yi-34B](https://huggingface.co/NousResearch/Nous-Hermes-2-Yi-34B) which has better commercial licenses, and bilingual support - More diverse and high quality data mixture - Dynamic high resolution ![image/png](https://cdn-uploads.huggingface.co/production/uploads/62441d1d9fdefb55a0b7d12c/FPshq08TKYD0e-qwPLDVO.png) ## Intended uses & limitations You can use the raw model for tasks like image captioning, visual question answering, multimodal chatbot use cases. See the [model hub](https://huggingface.co/models?search=llava-hf) to look for other versions on a task that interests you. ### How to use Here's the prompt template for this model: ``` "[INST] \nWhat is shown in this image? [/INST]" ``` You can load and use the model like following: ```python from transformers import LlavaNextProcessor, LlavaNextForConditionalGeneration import torch from PIL import Image import requests processor = LlavaNextProcessor.from_pretrained("llava-hf/llava-v1.6-mistral-7b-hf") model = LlavaNextForConditionalGeneration.from_pretrained("llava-hf/llava-v1.6-mistral-7b-hf", torch_dtype=torch.float16, low_cpu_mem_usage=True) model.to("cuda:0") # prepare image and text prompt, using the appropriate prompt template url = "https://github.com/haotian-liu/LLaVA/blob/1a91fc274d7c35a9b50b3cb29c4247ae5837ce39/images/llava_v1_5_radar.jpg?raw=true" image = Image.open(requests.get(url, stream=True).raw) # Define a chat history and use `apply_chat_template` to get correctly formatted prompt # Each value in "content" has to be a list of dicts with types ("text", "image") conversation = [ { "role": "user", "content": [ {"type": "text", "text": "What is shown in this image?"}, {"type": "image"}, ], }, ] prompt = processor.apply_chat_template(conversation, add_generation_prompt=True) inputs = processor(images=image, text=prompt, return_tensors="pt").to("cuda:0") # autoregressively complete prompt output = model.generate(**inputs, max_new_tokens=100) print(processor.decode(output[0], skip_special_tokens=True)) ``` ### Model optimization #### 4-bit quantization through `bitsandbytes` library First make sure to install `bitsandbytes`, `pip install bitsandbytes` and make sure to have access to a CUDA compatible GPU device. Simply change the snippet above with: ```diff model = LlavaNextForConditionalGeneration.from_pretrained( model_id, torch_dtype=torch.float16, low_cpu_mem_usage=True, + load_in_4bit=True ) ``` #### Use Flash-Attention 2 to further speed-up generation First make sure to install `flash-attn`. Refer to the [original repository of Flash Attention](https://github.com/Dao-AILab/flash-attention) regarding that package installation. Simply change the snippet above with: ```diff model = LlavaNextForConditionalGeneration.from_pretrained( model_id, torch_dtype=torch.float16, low_cpu_mem_usage=True, + use_flash_attention_2=True ).to(0) ``` ### BibTeX entry and citation info ```bibtex @misc{liu2023improved, title={Improved Baselines with Visual Instruction Tuning}, author={Haotian Liu and Chunyuan Li and Yuheng Li and Yong Jae Lee}, year={2023}, eprint={2310.03744}, archivePrefix={arXiv}, primaryClass={cs.CV} } ```