FlowiseAI (YC S23)

Single Sign On (SSO) is now available on Flowise Enterprise! This feature allows using a single identity provider (IdP), starting with Microsoft Entra ID https://lnkd.in/eUmKmt4t

Stripe recently released new SDK built for AI agents: https://lnkd.in/euy3dXFb The new era of financial agentic workflows is here. Here's an example workflow by one of our users in manufacturing industry: 1.) Upload invoice images 2.) LLM calculates total bill 3.) Payment link created with the calculated amount Excited to see what can we build with this!

Flowise v2.1.4 release 🥳 📈 Prometheus Grafana, OpenTelemetry Native support for Prometheus with Grafana and OpenTelemetry for metrics such as API requests, counts of flows/predictions. https://lnkd.in/eYb8v5bM 🦙 Ollama multi modal Process images, files and enable JSON mode ✨ Cerebras LLM Use Cerebras Systems lightning fast inference speed that can delivers 1,800 tokens/sec for Llama 3.1-8B and 450 tokens/sec for Llama 3.1-70B, 20x faster than NVIDIA GPU-based hyperscale clouds. 🛋️ Couchbase Vector Store Couchbase is the NoSQL cloud database platform for business-critical, AI-ready applications. Leverage their vector search capabilities for RAG apps ✍ Override Configuration ⚠️ BREAKING Change Due to security reason, override config is disabled by default. User must explicitly enable this in the Security tab. https://lnkd.in/eE6ATVfy 🗄️ Postgres, MySQL support for Agent Memory Agent Memory is a crucial piece when designing an agentflow. It remembers the state, checkpoints of the conversation. Now supporting PostgresQL and MySQL! 🌐 Global Proxy If you're running Flowise in an environment that requires a proxy, such as within an organizational network, you can configure Flowise to route all its backend requests through a proxy of your choice. https://lnkd.in/ei7U7mzm There are still lots of bugfixes and improvements contributed by community 🙌 For more details, take a look at the release: https://lnkd.in/eHFGfWqj

This is definitely an interesting use case of LLM!

Flowise v2.1.3 is now out and comes with a lot of goodies 🎃 📂 Full and RAG File Uploads - RAG file uploads embed and upsert to the vector store in real-time. - Full file uploads send entire files to the LLM, ideal for large-context models like Gemini and Claude. See how we upload a CSV and generate a summary graph using the full file upload feature 👇 🔥 New LLMs and Embeddings - Claude 3.5 Sonnet from Anthropic, Bedrock, Vertex - NVDIA Nemo Guardrails - JinaAI Embedding - Alibaba Tongyi - Cerebras Thank you to all the contributors! As always lots of bugfix and small improvement: https://lnkd.in/eJJatgHm

We're super excited to introduce Workspaces in Flowise ✨ Workspaces isolate resources among various teams, with Role-Based Access Control (RBAC) for permissions and access. Simply put, you can now invite others to collaborate on your flows/agents! https://lnkd.in/et9n6PPt

We released a small update v2.1.2 but its immensely useful 🌐 OpenAPI Toolkit This allows you to upload an OpenAPI 3.0 YAML spec file, and automatically convert each APIs into set of tools for function calling. From the demo video, we uploaded OpenAI API yaml file with 20 endpoints and have the LLM automatically picks and executes the tools required to answer user query. 💬 Follow up questions Generate follow up questions based on the previous conversation All this and more, including tons of bug fixes and UI/UX improvements, in the latest v2.1.2! https://lnkd.in/dsph3WMU

What if you could create your own OpenAI Assistant with Realtime API that can: 🌐 Browse the Web 📚 Search Files (RAG) 💻 Run Code Interpreter All using open source tools. Check it out

<<< An example of #Healthcare #Innovation Using #Generative #AI >>> With the recent code interpreter and file upload features from E2B, and FlowiseAI (YC S23), I built a #Diabetes #AI Advisor and Assistant equipped with #Data #Analytics Tools to aid patients, parents, and doctors in smartly monitoring, managing, and gaining more insights from diabetes data records. The uploaded data are upserted to its own #vector store, alongside its time-series data. This allows for comparisons to be made over longer time periods.  #HealthTech #MedicalAI #ArtificialIntelligence 

⭐ 30K GitHub Stars !!! Just over a year ago, we embarked on this journey, and today we’re thrilled to celebrate 30K stars and 15K forks on GitHub! We’re committed to continuously improving and becoming the best tool for building LLM apps 🚀 Thank you, open source community 🙌