Stability AI Launches Open Source Chatbot Stable Chat – InfoQ.com
Facilitating the Spread of Knowledge and Innovation in Professional Software Development
Thomas Dullien discusses how language design choices impact performance, how Google’s monorepo culture and Amazon’s two-pizza-team culture impact code efficiency, and why statistical variance is an enemy.
In this episode, Roi Ravhon, co-founder and CEO of Finout, sat down with InfoQ podcast co-host Daniel Bryant and discussed the emergence and industry adoption of FinOps. The talk covers topics such as the benefits of adopting FinOps, the typical journey of an organization interested in learning more about cloud costs, and a range of cultural and tooling required for a successful implementation.
Felix GV explains what derived data is, and dives into four major use cases which fit in the derived data bucket, including: graphs, search, OLAP and ML feature storage.
In this podcast Shane Hastie, Lead Editor for Culture & Methods spoke to Phil Alves about how leaders can enable developer effectiveness.
Courtney Nash discusses research collected from the VOID, challenging standard industry practices for incident response and analysis, like tracking MMTR and using RCA methodology.
See what’s next in Large Language Model (LLM) architectures used in tools like ChatGPT.
Learn what’s next in software from world-class leaders pushing the boundaries. Attend in-person or get video-only pass to recordings.
Discover new ideas and insights from senior practitioners driving change in software. Attend in-person.
Your monthly guide to all the topics, technologies and techniques that every professional needs to know about. Subscribe for free.
InfoQ Homepage News Stability AI Launches Open Source Chatbot Stable Chat
Aug 29, 2023 2 min read
by
Stability AI, makers of the image generation AI Stable Diffusion, recently launched Stable Chat, a web-based chat interface for their open-access language model Stable Beluga. At the time of its release, Stable Beluga was the best-performing open large language model (LLM) on the HuggingFace leaderboard.
Stable Beluga is based on the LLaMA foundation model released by Meta. The model is fine-tuned using a synthetic dataset generated by GPT-4. The largest Stable Beluga model contains 70B parameters and outperforms ChatGPT on several benchmarks, including AGIEval, which is based on several common examinations such as LSAT and SAT. To help evaluate Stable Beluga, Stability AI created the Stable Chat web interface to help users interact with the model and give feedback on its output. According the Stability AI,
As part of our efforts at Stability AI to build the world's most trusted language models, we've set up a research-purpose-only website to test and improve our technology. We will continue to update new models as our research progresses rapidly. We ask that you please avoid using this site for real-world applications or commercial uses.
The Stable Beluga models were inspired by a paper published by Microsoft on Orca, a fine-tuned version of LLaMA. In the paper, Microsoft described a technique called explanation tuning. Like instruction tuning, which has been used on many open LLMs recently, including ChatGPT and Vicuna, explanation tuning uses a dataset of example inputs and desired model outputs that are generated by a teacher. In the case of ChatGPT, the teachers are actual human users of the model. In contrast, for Orca and Stable Beluga, the explanation tuning dataset is generated by prompting GPT-4 to explain why it generated the output it did ("explain like I’m five.")
Stability AI created their own explanation tuning dataset of 600,000 examples—one-tenth the size of the Microsoft dataset. They then trained two versions of Stable Beluga: Stable Beluga 1, based on the 65B parameter original LLaMA model, and Stable Beluga 2, based on the 70B Llama 2 model. Both are released under a non-commercial license. Although the models achieved fourth and first place, respectively, on the leaderboard when they were released, the proliferation of LLaMA-based fine-tuned models has currently pushed Stable Beluga 2 out of the top ten, and Stable Beluga 1 even lower.
The models were released under a non-commercial license to encourage researchers to help iterate and improve on the technology, according to Stability AI. However, the company noted that this required resources that are "beyond the reach of everyday researchers," and decided to create the Stable Chat website. Users can create a free login or use a Google account to access the chat. The responses from the model can be up-voted, down-voted, or flagged; this user feedback will be used to help improve the model in the future.
Stability AI founder Emad Mostaque posted about the release on Twitter/X. One user replied that the model was "too cautious in giving factual information." Mostaque urged the user to give that feedback via the web interface.
Stability AI also recently announced that their LLMs will be used at an AI red-teaming event at DEF CON 31. This event is sponsored by the White House and features models from "Anthropic, Google, Hugging Face, Microsoft, NVIDIA, OpenAI, and Stability AI." The goal is to help identify risks and vulnerabilities in the models.
Writing for InfoQ has opened many doors and increased career opportunities for me. I was able to deeply engage with experts and thought leaders to learn more about the topics I covered. And I can also disseminate my learnings to the wider tech community and understand how the technologies are used in the real world.
I discovered InfoQ’s contributor program earlier this year and have enjoyed it since then! In addition to providing me with a platform to share learning with a global community of software developers, InfoQ’s peer-to-peer review system has significantly improved my writing. If you’re searching for a place to share your software expertise, start contributing to InfoQ.
I started writing news for the InfoQ .NET queue as a way of keeping up to date with technology, but I got so much more out of it. I met knowledgeable people, got global visibility, and improved my writing skills.
Becoming an editor for InfoQ was one of the best decisions of my career. It has challenged me and helped me grow in so many ways. We'd love to have more people join our team.
InfoQ seeks a full-time Editor-in-Chief to join C4Media’s international, always remote team. Join us to cover the most innovative technologies of our time, collaborate with the world’s brightest software practitioners, and help more than 1.6 million dev teams adopt new technologies and practices that push the boundaries of what software and teams can deliver!
A round-up of last week’s content on InfoQ sent out every Tuesday. Join a community of over 250,000 senior developers. View an example
We protect your privacy.
You need to Register an InfoQ account or Login or login to post comments. But there’s so much more behind being registered.
Get the most out of the InfoQ experience.
Allowed html: a,b,br,blockquote,i,li,pre,u,ul,p
Allowed html: a,b,br,blockquote,i,li,pre,u,ul,p
Allowed html: a,b,br,blockquote,i,li,pre,u,ul,p
A round-up of last week’s content on InfoQ sent out every Tuesday. Join a community of over 250,000 senior developers. View an example
We protect your privacy.
October 2-6, 2023.
Attend in-person. Or get a Video-Only Pass to watch recordings later.
QCon San Francisco International Software Conference returns this October 2-6. More than 1000 software professionals will join together and learn about the emerging trends they should pay attention to in 2023, how to adopt them, how to avoid pitfalls, and how to embrace the best practices.
Join the experience and get implementable ideas to shape your projects that last beyond the conference.
SAVE YOUR SPOT NOW
InfoQ.com and all content copyright © 2006-2023 C4Media Inc.
Privacy Notice, Terms And Conditions, Cookie Policy
