𝗛𝗮𝘆𝘀𝘁𝗮𝗰𝗸 𝗰𝗮𝗻 𝗻𝗼𝘄 𝘀𝗲𝗲 👀 The latest #haystack release adds a long-requested feature: 𝗶𝗺𝗮𝗴𝗲 𝘀𝘂𝗽𝗽𝗼𝗿𝘁! 📓 Notebooks in the comments This isn't just about passing images to an LLM. We built several features to enable practical multimodal use cases. 𝗪𝗵𝗮𝘁'𝘀 𝗻𝗲𝘄 🧠 Support for multiple LLM providers: OpenAI, Amazon Bedrock, Google Gemini, Mistral AI, NVIDIA, OpenRouter, Ollama and more 🎛️ Prompt template language to handle structured inputs, including images 📄 PDF and image converters 🧾 LLM-based extractor to pull text from images 🔍 Image embedders using CLIP-like models 🧩 Components to build multimodal RAG pipelines and Agents I had the chance of leading this effort with Sebastian Husch Lee, PhD (great collab). 📓 Below you can find two notebooks to explore the new features: • Introduction to Multimodal Text Generation • Creating Vision+Text RAG Pipelines (🖼️ image by Bilge Yücel) #largelanguagemodels #multimodality #opensource #rag #agents
Stefano Fiorucci, this update on Haystack is truly exciting. The integration of image support opens up a world of possibilities for multimodal applications. Collaborating with experts like Sebastian H. is an assurance of innovation. Looking forward to exploring those notebooks and all the creative potentials they unlock. #opensource