Pinned One Name. One Product Family. One Look 💙 We’re unifying the Haystack ecosystem at @deepset_ai under one name and a new logo, reflecting its role as a framework, a community, and the foundation of our enterprise platform. 👉 Read the announcement: haystack.deepset.ai/blog/announcin…
We are thrilled to announce the stable version of Haystack 2.0 🎉 We’ve been working on this for a while and now Haystack 2.0 has everything to help you implement composable LLM applications that are easy to use, customize, extend, optimize, evaluate, and deploy to production.
Prompt Optimization with Haystack and DSPy When building applications with LLMs, writing effective prompts is a long process of trial and error 🔄 Often, if you switch models, you also have to change the prompt 😩 What if you could automate this process? 💡 That's where DSPy
Do you like to try new open LLMs (like Qwen 2 😉)? In Haystack, open models are first-class citizens. 🤗 @huggingface is the home of open models and we are very keen to support their different libraries and solutions. For example, HuggingFaceLocalGenerator, based on
🦙 @Ollama lands in the Haystack ecosystem! Ollama is the equivalent of 🐳 Docker for LLMs: an easy way to pack and run quantized LLMs everywhere, even in cheap laptops (wo GPUs). This integration, driven by community demand, was also implemented by the community! Learn more
Cohere Rerank 3 + Haystack 2.0 @cohere just dropped new great models for reranking 👇 · Context length of 4k for significant improvement on longer documents · Multilingual coverage of 100+ languages · Improved latency You can use it right away in Haystack to improve your
🗣️ Introducing @AssemblyAI to the Haystack ecosystem! Check out the new integration with AssemblyAI models to effortlessly transcribe audio, generate summaries, or execute speaker diarization within your pipelines. 📯 Read the announcement: assemblyai.com/blog/announcin…🧑💻 Learn
Hello world 👋 We're new here. We're the team that develops Haystack, the open-source framework for LLM orchestration by @deepset_ai . Follow us for regular updates and news on Haystack.
👋 Say hello to BM42 🔎 BM42 is a new ranking algorithm implemented by @qdrant_engine to overcome the limitations of BM25 for hybrid RAG pipelines. It uses Transformer models for better quantification of term importance. See it in action with Haystack: github.com/deepset-ai/hay…
Get ready for the first-ever Advent of Haystack!! We're super excited to share that we are releasing Haystack 2.0-Beta, our first official release for what will become Haystack 2.0, which you can try out by completing 10 challenges we will be publishing throughout the month of
Hey, do you learn best by building? Join us for Hack Night, March 26th, at @github HQ in San Francisco. Featuring @weaviate_io , @jamdotdev , & @neosynccloud . 🙊 No talks, just LLM hackin' 🎁 earn prizes for completing coding jams register here:
llama.cpp is one of the first tools that comes to mind when running LLMs locally. It uses the efficient quantized GGUF format, dramatically reducing memory requirements and accelerating inference without a GPU! 🏎️ Thanks to @awinml , Haystack now has llama.cpp integration that
Turn your AI ideas into reality with @AndrewYNg and @tuanacelik in a new @DeepLearningAI course: "Building AI Applications with Haystack"! In one hour we'll take you from understanding Haystack’s building blocks: components and pipelines, to creating smart chat agents that
Feature appreciation post. Sometimes, code can get complex. So why not simply draw? ✍️🎨 With Haystack 2.0, pipelines will be very flexible. Sometimes, it's nice to quite literally look at them! An example of a RAG pipeline with a recentness ranker👇 Simply: pipeline.draw()
