Shale Protocol

Add LLMs to your apps via a simple API call.

Our beta is live!

Github:

[Shale Protocol @ShaleProtocol]

The wait is over. Shale-Serve API now supports inferences from the Llama2-13b-chat model by default.

Read the documentation to learn how you can seamlessly combine Shale-Serve API with @langchain.

https://t.co/B24gkqjrjX

[Joey DeBruin @joey_debruin]

It's launch day today — give this a boost to help >1000 builders that have shipped like hell over the past 4 weeks get their products out there into the world 🚀

Congrats V3! Next step, helping grants and opportunities flow to these builders 🫡

[Robo @robodottech]

it’s Build V3 launch day🚀🚀🚀

thousands of builders from all over the world came together over the past 4 weeks to build something new

congrats to everyone who built on #BackdropBuild this time around

builders do your thing and add your launches in the comments below 👇

[Shale Protocol @ShaleProtocol]

Thanks to those who responded to this poll. We ran the poll concurrently on Twitter and LinkedIn. Even though the limited number of responses won't give us any conclusive insights, they do help us test some ideas.🙃

[Shale Protocol @ShaleProtocol]

If you have developed a fine-tuned domain-specific #LLM with an open-source base model, would you be interested in monetizing the model per se? e.g., setting up an API service for recurring revenue.

Please leave your comments below.👇

[Shale Protocol @ShaleProtocol]

gm!

How did some of our users discover our product and community? 🥰

"I've been looking through numerous options to set up these LLM inferences for my research, but couldn't find any suitable platforms. Most of them are costly. I'm glad to now be a part of such a great community."

"😃Funny story. Since I couldn't find any platforms that were viable for me, I started reading through each and every plugin/integration of @langchain python library and eventually found this community."

https://t.co/bL1bz08CVX

[Shale Protocol @ShaleProtocol]

@Skruffster Thanks for the kind words, @Skruffster

[LangChain @LangChain]

⭐️LangChain Integrations⭐️

In this fast moving LLM landscape, we want to give everyone the power and flexibility of as many options as possible

But we also want to make it simple and easy to navigate

That's why we're launching an integrations hub

https://t.co/k1stYZA1KC

[Shale Protocol @ShaleProtocol]

With the same API key, anyone can access these four open-source LLMs with the Shale-Serve API.

[Shale Protocol @ShaleProtocol]

The power of open source lies in community co-creation. 👾

CodeT5-6B is part of a new family of open-source LLMs for code understanding and generation.

What would you be able to build when you combine codet5p-6b, Falcon-7b, and @langchain? 👩‍💻

https://t.co/g6YaUYUHr7

[Arena.ai @arena]

Introducing FastChat's OpenAI-compatible API server!

Many applications have been built on OpenAI APIs, but now you can effortlessly port them to use open-source alternatives without modifying the code.

Blog (with LangChain + Vicuna as an example):
https://t.co/tYJSvtOWjm

[Shale Protocol @ShaleProtocol]

See how we added integration of codet5p-6b to the FastChat API server.

https://t.co/eaWtmcfCKu

[Shale Protocol @ShaleProtocol]

We've made the user onboarding process as frictionless as possible.

Join our Discord and generate an API key.

No hassles with inputting a cc and free trials.

The free tier supports up to 1,000 daily requests per key.

https://t.co/vuFUm7ePX1

[Shale Protocol @ShaleProtocol]

Release updates: now, with the same API key, anyone can access these open-source LLMs with the Shale-Serve API:

vicuna-13b-v1.1 by @lmsysorg (cc: @lm_zheng)
codet5p-6b by @salesforce (cc: @ayueei )
falcon-7b & falcon-7b-instruct by @TIIuae

[Shale Protocol @ShaleProtocol]

The API is compatible with the openai-python library and @langchain. (cc: @hwchase17)

Three of these four models have a permissible license for commercial use.

Join our Discord and generate an API key. https://t.co/vuFUm7fnMz

[Shale Protocol @ShaleProtocol]

Release updates: now, with the same API key, anyone can access these open-source LLMs with the Shale-Serve API:

vicuna-13b-v1.1 by @lmsysorg (cc: @lm_zheng)
codet5p-6b by @salesforce (cc: @ayueei )
falcon-7b & falcon-7b-instruct by @TIIuae

[Harrison Chase @hwchase17]

Many startups focused on serving open source models have exposed APIs that are interoperable with OpenAI

This makes it easy for anyone to try them out by just changing the API base

We added better support for this in @langchain in the most recent release (h/t Yuze Ma) 🧵

[Shale Protocol @ShaleProtocol]

@KagabeAI @langchain @streamlit @jasonzhou1993 🔥❤️

[Shale Protocol @ShaleProtocol]

PSA: It looks like our friends at @langchain have recently reorganized their docs. The section for LLM integrations has been moved to:
https://t.co/4OqXWPI6nh

Find the documentation for the Shale-Serve API here: https://t.co/WvRnzGn1Xd

@hwchase17

[Shale Protocol @ShaleProtocol]

We are working on integrating a few more open-source LLMs into the Shale-Serve API. We'll make an announcement as soon as they are available for production. 🔥

@langchain @lmsysorg

[Shale Protocol @ShaleProtocol]

@35mmChemistry