DiamondGotCat/
Zeta-1:Q6_K

26 3 months ago

Zeta-1 is a lightweight 405M LLM for conversational tasks.

3 months ago

1b3bd90ec2f7 · 335MB ·

model
gptneox
·
405M
·
Q6_K
params
{ "stop": [ "</ASSISTANT>" ] }
template
{{ if .System }}{{ .System }}{{ end }} {{ if .Prompt }}<USER>{{ .Prompt }}</USER>{{ end }} <ASSISTAN

Readme

Next Version has Released! v2

Zeta-1-Model-Image.png

Zeta-1

Zeta-1 is a small yet expressive open LLM with 405M parameters.

Zeta-1 is optimized for: - Lightweight environments - Fast inference speeds - Simple fine-tuning workflows - Educational and experimental use cases

Key Features

  • 🚀 ~405M parameters = fast and light
  • 🤖 Good performance on common questions, tech topics, and general reasoning
  • 🧩 Easily extendable and customizable with Zeta-Tool

Quantization

This model is available in GGUF format via Zeta-1-GGUF thanks to the amazing contribution by @mradermacher. It is compatible with Ollama and llama.cpp.

Credits

Example Output

What is Xcode?
Zeta-1: “Xcode is a development environment and toolset for building iOS apps…”

License

MIT License – free to use, remix, and improve.