MiniGPT-4 Reviews and details

Social recommendations and mentions

• Multimodal LLM for infographics images

  Isn't there only two open multimodal LLMs, LLaVA and mini-gpt4? Source: 10 months ago

• Upload a photo of your meal and get roasted by ChatGPT

  So we use MiniGPT-4 for image parsing, and yep it does return a pretty detailed (albeit not always accurate) description of the photo. You can actually play around with it on Huggingface here. Source: 12 months ago

• Upload a photo of your meal and get roasted by ChatGPT

  We use MiniGPT-4 first to interpret the image and then pass the results onto GPT-4. Hopefully, once GPT-4 makes its multi-modal functionality available, we can do it all in one request. Source: 12 months ago

• Give some love to multi modal models trained on censored llama based models

  But I would like to bring up that there are some multi models(llava, miniGPT-4) that are built based on censored llama based models like vicuna. I tried several multi modal models like llava, minigpt4 and blip2. Llava has very good captioning and question answering abilities and it is also much faster than the others(basically real time), though it has some hallucination issue. Source: 12 months ago

• Where can buy an openai account with GPT-4 access?

  Https://minigpt-4.github.io/ <-- free image recognition, although not powered by true GPT-4. Source: about 1 year ago

• Created a discord bot that understand images (based on MiniGPT-4 Vicuna model)

  I used the one from MiniGPT-4 paper: https://minigpt-4.github.io. Source: about 1 year ago

• Introducing Replicable.Art: A Curated SD Gallery and Competition Platform

  Yeah that will probably be exactly the way it will be solved, the building bricks for that solution are already there too with CLIP based developments like PEZ and MiniGPT. Source: about 1 year ago

• I've built an AI (Dall-E) powered chrome extension to generate similar images

  If you're absolutely dead set on using Dall-E, then you could pull the image through MiniGPT-4, LLaVA, BLIP-2 or OpenFlamingo to extract the image's visual content and generate an image with Dall-E 2 based on that. Source: about 1 year ago