mrjacktung / mradermacher-llamax3-8b-alpaca-gguf

LLaMAX3-8B is a multilingual language base model, developed through continued pre-training on Llama3, and supports over 100 languages.

8B

32 Pulls Updated 2 months ago

2 months ago

f37dc3ee60c7 · 4.9GB

model
llama
·
8.03B
·
Q4_K_M
template
{{ .System }} ### Input: {{ .Prompt }} ### Response:

Readme

mradermacher/LLaMAX3-8B-Alpaca-GGUF


- Paper: LLaMAX: Scaling Linguistic Horizons of LLM by Enhancing Translation Capabilities Beyond 100 Languages.
- Link: https://arxiv.org/pdf/2407.05975
- Repository: https://github.com/CONE-MT/LLaMAX/
- Example: https://github.com/Mr-Jack-Tung/JARVIS-Chatbot-a-simple-RAG-with-PDF-files

How To Create Custom Ollama Models From HuggingFace ( GGUF ) file
step 1: Go to
https://huggingface.co/mradermacher/LLaMAX3-8B-Alpaca-GGUF

step 2: Download GGUF file
Choice: Q4_K_M | 5.0GB | fast | recommended

step 3: Create Modelfile
echo ‘FROM “./LLaMAX3-8B-Alpaca.Q4_K_M.gguf”\nTEMPLATE “{{ .System }}\n### Input:\n{{ .Prompt }}\n### Response:”’ >> Modelfile

step 4: Login your Ollama account (eg. mrjacktung)
My models ~> New ~> create new space with name: mradermacher-llamax3-8b-alpaca-gguf

step 5: Create repository
ollama create -f Modelfile mrjacktung/mradermacher-llamax3-8b-alpaca-gguf
ollama push mrjacktung/mradermacher-llamax3-8b-alpaca-gguf

step 6: Testing
ollama run mrjacktung/mradermacher-llamax3-8b-alpaca-gguf

Thanks to:
- Michael Radermacher: https://huggingface.co/mradermacher/LLaMAX3-8B-Alpaca-GGUF
- Data Science Basics: Ollama, How To Create Custom Models From HuggingFace (GGUF) _ https://www.youtube.com/watch?v=TFwYvHZV6j0