lgrammel

In case you're using JS/TS, I've been working on an abstraction layer for integrating AI models applications, unifying the API for common operations such as text streaming, object generation, and tool usage: https://github.com/lgrammel/modelfusion

Thanks, I appreciate it! Good luck with the project!

My expectation would have been a fork, or a repo thats started with the full commit history.

Thanks!

So, you essentially took my project ( https://github.com/rubberduck-ai/rubberduck-vscode ), and instead of contributing to it or forking it, you recommitted all the code under your own name, with a few small tweaks.

To be fair, I saw that you give credit, but it's kind of disappointing still given that prob 98% of the project is the work of the Rubberduck contributors. I know Rubberduck is not that active any more, but it's still disappointing.

Hey!

I've created "StoryTeller", a multi-modal app that quickly generates audio stories for pre-school kids.

StoryTeller is built with the following libraries:

- Fastify - Next.js - shadcn/ui - ModelFusion - Zod

The following AI APIs are used:

- OpenAI (story generation, embeddings) - Eleven Labs (tts) - Lmnt (tts) - thanks for sponsoring the project with credits! - Stability (images)

https://github.com/lgrammel/storyteller

StoryTeller is an exploratory web application that creates short audio stories for pre-school kids.

It used speech-to-text, llms, text-to-speech, embeddings and image generation.

StoryTeller is built with the following libraries:

ModelFusion Fastify Next.js shadcn/ui Zod

The following AI APIs are used:

OpenAI Eleven Labs Lmnt Stability

I think it is more nuanced. This article for example contains examples that suggest otherwise if you want to increase quality (which is a major concern when putting things in production):

https://www.anyscale.com/blog/fine-tuning-llama-2-a-comprehe...

Thanks :) I’ll post here when I have larger updates.

Yes, good point, JS/TS is definitely behind Python. That might explain some of it.

I expect most models to become multi modal in the future and am building towards. A lot of the core logic of agents will nevertheless be text based imo, so that’s a central piece, but I already added text to image and speech to text, and plan to add text to speech next.

I’m working on ModelFusion, a TypeScript library for working with AI models (llm, image, etc.)

https://github.com/lgrammel/modelfusion

It is only getting limited traction so I’m wondering if I’m missing something fundamental with the approach that I’m taking.

If you work with JS or TS, check out this alternative that I've been working on:

https://github.com/lgrammel/modelfusion

It lets you stay in full control over the prompts and control flow while make a lot of things easier and more convenient.

You can also spin up the llama.cpp server ( https://github.com/ggerganov/llama.cpp/tree/master/examples/... ) and e.g. use ModelFusion to generate text, stream text, tokenize, etc. : https://modelfusion.dev/integration/model-provider/llamacpp

llama.cpp has a server with a REST API that you can use: https://github.com/ggerganov/llama.cpp/tree/master/examples/...

ModelFusion is a library for building AI apps, chatbots, and agents. It provides abstractions for AI models, vector indices, and tools.

- Type inference and validation: ModelFusion uses TypeScript and Zod to infer types wherever possible and to validate model responses.

- Flexibility and control: AI application development can be complex and unique to each project. With ModelFusion, you have complete control over the prompts and model settings, and you can access the raw responses from the models quickly to build what you need.

- No chains and predefined prompts: Use the concepts provided by JavaScript (variables, functions, etc.) and explicit prompts to build applications you can easily understand and control. Not black magic.

- More than LLMs: ModelFusion supports other models, e.g., text-to-image and voice-to-text, to help you build rich AI applications that go beyond just text.

- Integrated support features: Essential features like logging, retries, throttling, tracing, and error handling are built-in, helping you focus more on building your application.

Oh yeah, never thought that - was just frustrated at the lack of success in my marketing efforts.

Thanks for the feedback, good points!

Author here - thanks for the mention! Re publicizing: I can never get any attention for my projects, no matter where I post them. Sites like this one (where I tried posting the project multiple times) feel very gamed to me at this point.

If you’re using JS/TS and want to have a nicer API for LLM calls (and a bit more), check out https://github.com/lgrammel/ai-utils.js

I'm developing Rubberduck, an opensource AI chatbot and tool for Visual Studio Code. It can edit code, explain it, generate tests, diagnose errors, or talk about software development.

I plan to release a v1.0 in the coming days and I'm looking for feedback on what to improve.

Hey :)

As part of a larger project that I'm working on, I've created a playground where you can explore some newer (and older) parts of JS, e.g. the ?? and ?. operators, and see how existing code could be changed to use them.

Would love to hear feedback, in particular since this idea is pretty early :)