Welcome, Cognitive Revolution fans!
Intuitive visualizations allow you to interrogate every step of your LLM program.
Created on August 29|Last edited on September 5
Comment
Introducing Weave from Weights & Biases
Weave is a game-changing toolkit for developers working on generative AI applications. It offers a seamless way to track, evaluate, and debug LLM-based projects.
You can use Weave to:
- Log and debug language model inputs, outputs, and metadata
- Build rigorous, apples-to-apples evaluations for language model use cases
- Organize all the information generated across the LLM workflow, from experimentation to evaluations to production
Check out the full capabilities of Weave
🧑💻 All you need are 3 lines of code
Get started by decorating your Python functions with @weave.op:
import weaveweave.init("cognitive-revolution")@weave.op()def sum_nine(value_one: int):return value_one + 9@weave.op()def multiply_two(value_two: int):return value_two * 2@weave.op()def main():output = sum_nine(3)final_output = multiply_two(output)return final_output
You can try Weave out in this interactive Colab:
♾️ Weave supports any use case
RAG applications, image generation, benchmarking, Weave can support you in any enterprise, academic or side project and here are some projects powered by Weave:
Automated PDF summarization of arXiv papers with Claude 3.5 Sonnet and W&B Weave
Learn how to create an automated PDF summarization system for arXiv papers using Anthropic's API and W&B Weave using Chain Of Density.
Prompt upsampling for diffusion models
This article shows the implementation of an LLM-assisted prompt upsampling strategy to improve the quality of images generated by Stable Diffusion.
How to optimize LLM workflows using DSPy and W&B Weave
Learn how to use DSPy teleprompters and Weave to automatically optimize prompting strategies for causal reasoning
Is the new Cerebras API the fastest LLM service provider?
Let's compare five different Llama 70B providers and run some benchmarks. We'll be looking at Cerebras, Groq, Together, Fireworks, and Octo.
🤝 Weave lives where you work
Weave offers integrations with many language model APIs and LLM frameworks to streamline tracking and evaluation:
Stay focused on iterating on your prompts and models with whichever API or LLM framework you know and love - Weave has integrations for OpenAI, Anthropic, Mistral.ai, LlamaIndex, Cohere and more.
See our integrations
🧑🎓 Explore Weave and build LLM apps with our free courses
Our courses will give you a theory and code to dive into the area of your interest. Our LLM courses range from short, problem focused courses to a more advanced deeper dives. Each course is equipped with theory, code and led by industry experts.
Check out all our courses (and enroll for free!)
Trusted by over 200,000 machine learning practitioners
Weights & Biases is trusted as the machine learning system of record by organizations across the work; from the largest deep learning research labs to autonomous driving companies and from pharmaceutical companies working on drug discovery to financial institutions. And with SOC2 certification and both cloud and local deployment options, Weights & Biases will meet your teams' security needs.
Try Weights & Biases today
Add a comment