V7 Go uses generative AI to automate tasks and document processing reliably and at scale, allowing companies to reduce the strain of back-office work and focus on what really matters. V7 Go turns images and documents into structured data. Check our latest release and new features here: https://www.v7labs.com/go/summer-release

Hey Product Hunt 👋 We built V7 Go because multimodal foundation models are good enough to solve most of the tasks we complete at work today, but aren’t easy to run at scale and with reliable results. We believe AI should free us from back-office and administrative tasks to give us time to focus on creative and interpersonal work. About 40% of the work we’re doing on our laptops can be done just as well by frontier models if configured correctly and with the right context, so we built a product that lets you intuitively configure AI models and run them at scale on any task involving documents, video, or text. Here’s a few things it can do out of the box: ✅ Extract information from large, complex documents ✅ Qualify and categorize inbound messages, such as contact forms, and perform actions via Zapier. ✅ Classify text, emails, or images in the order of millions. ✅ Compare the performance of different LLMs across the same task. As an AI-powered database, V7 Go can be used to solve almost anything, and upon launch we’ve focused a lot of its optimizations on document processing, form understanding, and data extraction, as tasks that nobody really likes doing, but are still done in massive quantities worldwide. 💡 The idea came to us when working with a client that analyzes 300,000 scanned pages of paper per day (!) occupying thousands of human hours. We not only needed a way to use LLMs like GPT-4 reliably enough to outperform expert human analysts, but also in a way that scaled past the 1:1 chat UI, so we developed V7 Go to resemble a smart spreadsheet where every cell hosts an LLM. There is a massive workflow challenge in generative AI, where its ability to succeed relies on being able to source similar examples of a task, and understand a company’s context, so it can be more accurate well-tuned than baseline GPT-4, Claude 3, or Gemini. V7 Go abstracts away a lot of configuration burdens on the user, but has a lot going on under the hood and is built to integrate with your (internal or external) product via the API. Here’s what makes it special: 🔬 Files are turned into mini repositories where an agent can analyze text, images, tables, and reflect on how to find the right info. We call this Index Knowledge 🔬 You can control the options a model spits out using Select Properties, so it won’t just ramble on with text. This is great for integrations. 🔬 Select Properties allow you to subdivide projects into Workflows, where each stage has an AI perform a unique set of operations. 🔬 Visual grounding (this one is in limited release) lets you see where each piece of info the LLM came up with is found in the source data, even if it’s an image. 🔬 Scales up to 10 million cells per project, that’s a lot of tokens! There’s more to come, and we’d love to hear your first impressions. ➡️ Special for PH: ScoutHunt10 for 20% off V7 Go Pro. You can try it for free now, we’d love your feedback and welcome any questions!

@albertorizzoli сongratulations on the release of V7 Go! Processing documents and extracting information from them frees up more time!

@albertorizzoli v7 looks amazing! I would love to take a peek under the hood. Thanks for sharing this product with the PH community, I’m excited to play around with various model options

@albertorizzoli awesome product V7 team, congratulations on the launch!

@albertorizzoli congratulations on the launch of V7 Go. I like the homepage and the UI/UX-Design. I can't wait to try it out!

@albertorizzoli Amazing product!

congratulations on the launch. I like the ability to choose different models. Any plans of integrating customs models, like with Huggingface?

@sanjuruk We're prioritizing multimodal LLMs at the moment but exploring the integration of HF models as tools that LLMs can call. For example if you need to run a bespoke classifier or detector as a workflow step. If there's any use case you have in mind let me know.

@albertorizzoli exciting, I'll let you know

@sanjuruk hey Sanju! Thank you for your comment. Yes, we are considering several ways of integrating external models into V7 Go, and integrating with Huggingface is one of them. Which Huggingface models would you use if we had said integration?

V7 Go's approach to generative AI for task automation seems incredibly efficient. Scaling this must have been a considerable endeavor. I’m curious about the key challenges you encountered when optimizing for different types of documents and images? In the development of Checky Bear, we're similarly focused on scaling a solution that helps website owners automate their security and stability, key to maintaining an effective online presence.

@albertorizzoli We truly appreciate the complex intricacies you've shared. The dedication towards tackling diverse document types and the pursuit of R&D for innovative solutions like video understanding is commendable. Thanks for considering to check out Checky Bear! We are excited to potentially learn from each other's journey in tech. Here's to pushing the boundaries of what's possible! 🚀🛡️

AIxBlock surpasses V7 Labs by offering on-premise instant self-hosting, giving companies complete privacy control over their data and AI models. While V7 Labs relies on cloud-based services, AIxBlock connects users to a decentralized marketplace of on-demand, diverse compute resources, providing cost savings of up to 90%

Truly revolutionary, the interface is incredibly easy to use and already automates one of my chore tasks at work! I love that I can choose from multiple LLMs and integrate them into a single interface.