The tech community is abuzz with the launch of several new tools and features aimed at enhancing the development and evaluation of generative AI applications. A new side-by-side mode has been introduced in a playground environment for comparing models, particularly beneficial for assessing fine-tuned models against their base versions. This feature, part of a broader update that includes weights and biases integration, the ability to save multiple checkpoints per training job, and a side-by-side comparison view in the API playground, received a special shout-out to @slessans and @karoliskosas for its launch. Additionally, a new tool named Weave, developed by @weights_biases, has been announced, designed to track and evaluate large language model (LLM) applications throughout their development lifecycle. Weave enables developers to log and version LLM interactions and surrounding data, experiment with prompting techniques, model changes, and parameters, and perform evaluations. OpenAI customers engaged in fine-tuning models can now log their data directly into Weave, marking a significant collaboration in the field. The introduction of Weave to the public cloud signifies a new era for Generative AI app development.
☁️ Weave is now live for all on the public cloud, introducing a new era for Generative AI app development. Learn more here → https://t.co/UXKcZIHLUu https://t.co/fnEij7Ij8r
Exciting news: @OpenAI customers that are fine tuning models can now log their data directly into @weights_biases - so far the only 3rd party company that seems to do this. Kudos to @l2k and team ! https://t.co/Bgj6BrPmwq
my favorite feature from today's release is the new side-by-side comparison UI. really fun to see how different models respond to the same prompt. shout-out to @slessans and @karoliskosas for launching this! 🚀 https://t.co/qCo51RMqs7
Introducing Weave by @weights_biases I'm so excited that we can share this finally! 🥳 If you're building with LLMs, you can use Weave for tracing and evals Get started by adding `weave.op()` to function definitions: https://t.co/wNs03QLizm https://t.co/xi09BBaK4W
I'm very excited to announce Weave, our new tools to track and evaluate your LLM apps. Use Weave to: 🍩log and version LLM interactions and surrounding data, from development to production 🍩experiment with prompting techniques, model changes, and parameters 🍩evaluate your… https://t.co/P6SnmEGPUi
new from fine-tuning: - weights and biases integration - saving multiple checkpoints per training job - a side by side comparison view in the API playground - and more! https://t.co/apoUsJONqz
just pushed a new side-by-side mode in our playground for comparing models. especially useful for vibe-checking your fine-tuned models against the base. https://t.co/fmDATe1gFV https://t.co/R8MGhgZNKk