Instructions to use xesutr/aom-news-4b with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- llama-cpp-python
How to use xesutr/aom-news-4b with llama-cpp-python:
# !pip install llama-cpp-python from llama_cpp import Llama llm = Llama.from_pretrained( repo_id="xesutr/aom-news-4b", filename="aom-news-4b.BF16-mmproj.gguf", )
llm.create_chat_completion( messages = "No input example has been defined for this model task." )
- Notebooks
- Google Colab
- Kaggle
- Local Apps
- llama.cpp
How to use xesutr/aom-news-4b with llama.cpp:
Install from brew
brew install llama.cpp # Start a local OpenAI-compatible server with a web UI: llama-server -hf xesutr/aom-news-4b:BF16 # Run inference directly in the terminal: llama-cli -hf xesutr/aom-news-4b:BF16
Install from WinGet (Windows)
winget install llama.cpp # Start a local OpenAI-compatible server with a web UI: llama-server -hf xesutr/aom-news-4b:BF16 # Run inference directly in the terminal: llama-cli -hf xesutr/aom-news-4b:BF16
Use pre-built binary
# Download pre-built binary from: # https://github.com/ggerganov/llama.cpp/releases # Start a local OpenAI-compatible server with a web UI: ./llama-server -hf xesutr/aom-news-4b:BF16 # Run inference directly in the terminal: ./llama-cli -hf xesutr/aom-news-4b:BF16
Build from source code
git clone https://github.com/ggerganov/llama.cpp.git cd llama.cpp cmake -B build cmake --build build -j --target llama-server llama-cli # Start a local OpenAI-compatible server with a web UI: ./build/bin/llama-server -hf xesutr/aom-news-4b:BF16 # Run inference directly in the terminal: ./build/bin/llama-cli -hf xesutr/aom-news-4b:BF16
Use Docker
docker model run hf.co/xesutr/aom-news-4b:BF16
- LM Studio
- Jan
- Ollama
How to use xesutr/aom-news-4b with Ollama:
ollama run hf.co/xesutr/aom-news-4b:BF16
- Unsloth Studio new
How to use xesutr/aom-news-4b with Unsloth Studio:
Install Unsloth Studio (macOS, Linux, WSL)
curl -fsSL https://unsloth.ai/install.sh | sh # Run unsloth studio unsloth studio -H 0.0.0.0 -p 8888 # Then open http://localhost:8888 in your browser # Search for xesutr/aom-news-4b to start chatting
Install Unsloth Studio (Windows)
irm https://unsloth.ai/install.ps1 | iex # Run unsloth studio unsloth studio -H 0.0.0.0 -p 8888 # Then open http://localhost:8888 in your browser # Search for xesutr/aom-news-4b to start chatting
Using HuggingFace Spaces for Unsloth
# No setup required # Open https://huggingface.co/spaces/unsloth/studio in your browser # Search for xesutr/aom-news-4b to start chatting
- Docker Model Runner
How to use xesutr/aom-news-4b with Docker Model Runner:
docker model run hf.co/xesutr/aom-news-4b:BF16
- Lemonade
How to use xesutr/aom-news-4b with Lemonade:
Pull the model
# Download Lemonade from https://lemonade-server.ai/ lemonade pull xesutr/aom-news-4b:BF16
Run and chat with the model
lemonade run user.aom-news-4b-BF16
List all available models
lemonade list
llm.create_chat_completion(
messages = "No input example has been defined for this model task."
)📰 aom-news-4b (Article/Analyst Optimized Model)
This repository contains aom-news-4b, a specialized large language model fine-tuned on Google Gemma 4B for structured news analysis. Its primary function is to generate a concise Summary, Sentiment Score, and a Numerical Score for financial and cryptocurrency articles.
The model is optimized for CPU and low-resource environments, offering a highly efficient solution for local inference and edge computing applications.
📄 Model Details & Limitations
| Detail | Value |
|---|---|
| Base Model | Google Gemma 4B (Fine-Tuned) |
| Parameter Count | ~4 Billion |
| Primary Task | Structured News Analysis & Sentiment Scoring |
| Optimal Use | Local Inference on resource-constrained devices (CPU-friendly) |
| License | Gemma Terms of Use |
⚠️ IMPORTANT LIMITATION: This model has been aggressively fine-tuned for a specific structural analysis task. It is not suitable for general conversation and will attempt to force all inputs into its defined output format.
⚖️ License
This model is a derivative work of the Gemma family of models. It is licensed under the Gemma Terms of Use. Please review the official Gemma Terms of Use provided by Google before using or redistributing this model.
💾 Available Model Formats
The model is provided in multiple formats for maximum deployment flexibility:
1. Safetensors (High-Precision)
- Use Case: Ideal for further fine-tuning or for deployment on high-performance servers/GPUs where maximum accuracy is required.
2. GGUF (Quantized for CPU/Edge)
- File Example:
aom-news-4b.Q8_0.gguf - Use Case: Recommended for Ollama, Llama.cpp, or any CPU/resource-constrained deployment due to its small size and high inference speed on standard hardware.
Example Input (API Payload or CLI)
Analyze the following news article and provide a summary, sentiment analysis, and a score.
TITLE: Maximizing Profit in a Volatile Market: How Vest Mining Transforms Crypto Uncertainty into Sustainable Growth BODY: Cryptocurrency presents amazing prospects as well as great market volatility. Vest Mining excels within this framework. With the use of sophisticated technology, intelligent automation, and a focus on increased stability, Vest Mining helps clients translate crypto volatility into reliable and sustainable income streams. CATEGORIES: BUSINESS, CRYPTOCURRENCY
Expected Output
The model will respond with the structured analysis:
SUMMARY: The article discusses how Vest Mining leverages cryptocurrency's inherent volatility to generate consistent revenue for its customers through technological advancements and automated processes, positioning it favorably amidst fluctuating crypto markets. SENTIMENT: POSITIVE SCORE: 0.9
- Downloads last month
- 31
# !pip install llama-cpp-python from llama_cpp import Llama llm = Llama.from_pretrained( repo_id="xesutr/aom-news-4b", filename="", )