LangChain(@LangChainAI) 's Twitter Profile Photo

❓How to get models to generate structured output?

JSONFormer (by rahul) and RELLM (by Matt Rickard) are two novel approaches for this, now with (experimental) integrations to LangChain

JSONFormer Integration python.langchain.com/en/latest/modu…

RELLM Integration python.langchain.com/en/latest/modu…

❓How to get models to generate structured output?

JSONFormer (by @rahulgs) and RELLM (by @mattrickard) are two novel approaches for this, now with (experimental) integrations to LangChain

JSONFormer Integration python.langchain.com/en/latest/modu…

RELLM Integration python.langchain.com/en/latest/modu…
account_circle
LangChain(@LangChainAI) 's Twitter Profile Photo

👐 OpenLM API 👐

OpenLM by Matt Rickard is a lightweight package that lets you interface with multiple model providers using the OpenAI API.

Try out the new OpenLM model in 🦜🔗 and see for yourself!

Docs: python.langchain.com/en/latest/modu…

👐 OpenLM API 👐

OpenLM by @mattrickard is a lightweight package that lets you interface with multiple model providers using the OpenAI API.

Try out the new OpenLM model in 🦜🔗 and see for yourself!

Docs: python.langchain.com/en/latest/modu…
account_circle
Will(@FinanceYF5) 's Twitter Profile Photo

在浏览器中完全运行 LLM,使用简单的headless React hook useLLM() 。作者:Matt Rickard 5月16日推出

Demo: chat.matt-rickard.com
win95界面太酷了,点击即可试用
为什么做成win95?作者说:Because the browser is the OS 并且有一套完整理论
GitHub: github.com/r2d4/react-llm

Thread:

account_circle
swyx(@swyx) 's Twitter Profile Photo

interesting mini essay from Matt Rickard on where the 'AI Engineer' should sit in the org

(shudder... reminds me of the endless 'Where should devrel sit?' conversations I just exited)

I suspect this is going to happen in two ways - 'center of excellence' type models where the

interesting mini essay from @mattrickard on where the 'AI Engineer' should sit in the org 

(shudder... reminds me of the endless 'Where should devrel sit?' conversations I just exited)

I suspect this is going to happen in two ways - 'center of excellence' type models where the
account_circle
Mostly Borrowed Ideas(@borrowed_ideas) 's Twitter Profile Photo

'Despite advances in technology and increased spending, the number of new drugs approved per billion dollars spent on research and development has halved approximately every nine years since the 1950s.' Matt Rickard

'Despite advances in technology and increased spending, the number of new drugs approved per billion dollars spent on research and development has halved approximately every nine years since the 1950s.' @mattrickard
account_circle
Brian Chambers(@BriChamb) 's Twitter Profile Photo

Really brilliant take from Matt Rickard on the parallels between language and software interfaces... and the possible emergence of some missing 'sentence modes' (interrogative, exclamatory) due to LLMs specifically and AI more generally.

blog.matt-rickard.com/p/imperative-d…

account_circle
Ethan ⛰️(@ethansteininger) 's Twitter Profile Photo

Matt Rickard Love it! In an effort to make each easier to use, I've indexed each into a GPT-powered knowledge hub:

collie.ai/rickard

ask questions like 'how do i use codegen' or 'what is polyglot'

feedback welcome!

account_circle
Davis Treybig(@TreybigDavis) 's Twitter Profile Photo

Hosting a foundation model builders meetup in Palo Alto in a few weeks: lu.ma/2ekxy7yw. It will be 'Show & Tell' style with a series of lightning talks - some of the speakers we have lined up include Jerry Liu Matt Rickard Devin Stein Foundry Tristan Zajonc

account_circle