overview for AnonymousD3vil

Is it just me or is prompt engineering basically useless with smaller models? in c/localllama@poweruser.forum

[–] AnonymousD3vil@alien.top 1 points 10 months ago

I've had success with 7b Llama2 for multiple prompt scenarios. Make sure you are defining the objective clearly.

At first after reading your post, I thought you're talking about something even smaller (phi-1/tiny llama).

1

LLM.js v1.0.2 is out with support for GGUF/GGML format and Model Playground (alien.top)

submitted 10 months ago by AnonymousD3vil@alien.top to c/localllama@poweruser.forum

1 comments fedilink

🚀 Exciting News! 🚀

Thrilled to announce the release of LLM.js v1.0.2! 🎉✨

🌐 LLM.js lets you play around with language models right in your browser, thanks to WebAssembly.

In this latest release, here's what's in store:

1️⃣ Expanded Format Support: Now GGUF/GGML formats are fully supported, thanks to the latest llama.cpp patch! 🦙 This opens up doors for various models like Mistral, Llama2, Bloom, and more!

2️⃣ Playground Fun: Explore and test different models seamlessly in playground demo! 🎮💬 Even from HF.

Feel free to check it out and share your thoughts! 🚀

LLM.js Playground: https://rahuldshetty.github.io/ggml.js-examples/playground.html

LLM.js: https://rahuldshetty.github.io/llm.js/

https://i.redd.it/ge16xwia512c1.gif

Any Good Speech-to-Speech Voice Changer Models? in c/localllama@poweruser.forum

[–] AnonymousD3vil@alien.top 1 points 10 months ago

You could do a audio transcription then TTS to achieve the similar results with whisper and coqui-ai TTS models.

https://github.com/coqui-ai/TTS