🗃️ Evaluation
4 items
📄️ Debugging
If you're building with LLMs, at some point something will break, and you'll need to debug. A model call will fail, or the model output will be misformatted, or there will be some nested model calls and it won't be clear where along the way an incorrect output was created.
🗃️ Deployment
1 items
🗃️ LangSmith
1 items
📄️ Model Comparison
Constructing your language model application will likely involved choosing between many different options of prompts, models, and even chains to use. When doing so, you will want to compare these different options on different inputs in an easy, flexible, and intuitive way.