Instructions to use awels/maximusLLM-3b-128k-gguf with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use awels/maximusLLM-3b-128k-gguf with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="awels/maximusLLM-3b-128k-gguf",
	filename="maximusLLM-phi3-128k-3b-v0.1.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 awels/maximusLLM-3b-128k-gguf with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf awels/maximusLLM-3b-128k-gguf
# Run inference directly in the terminal:
llama-cli -hf awels/maximusLLM-3b-128k-gguf

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf awels/maximusLLM-3b-128k-gguf
# Run inference directly in the terminal:
llama-cli -hf awels/maximusLLM-3b-128k-gguf

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 awels/maximusLLM-3b-128k-gguf
# Run inference directly in the terminal:
./llama-cli -hf awels/maximusLLM-3b-128k-gguf

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 awels/maximusLLM-3b-128k-gguf
# Run inference directly in the terminal:
./build/bin/llama-cli -hf awels/maximusLLM-3b-128k-gguf

Use Docker

docker model run hf.co/awels/maximusLLM-3b-128k-gguf

LM Studio
Jan
Ollama
How to use awels/maximusLLM-3b-128k-gguf with Ollama:
```
ollama run hf.co/awels/maximusLLM-3b-128k-gguf
```

Unsloth Studio new

How to use awels/maximusLLM-3b-128k-gguf 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 awels/maximusLLM-3b-128k-gguf 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 awels/maximusLLM-3b-128k-gguf to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for awels/maximusLLM-3b-128k-gguf to start chatting

Docker Model Runner
How to use awels/maximusLLM-3b-128k-gguf with Docker Model Runner:
```
docker model run hf.co/awels/maximusLLM-3b-128k-gguf
```

Lemonade

How to use awels/maximusLLM-3b-128k-gguf with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull awels/maximusLLM-3b-128k-gguf

Run and chat with the model

lemonade run user.maximusLLM-3b-128k-gguf-{{QUANT_TAG}}

List all available models

lemonade list

Maximus Model Card

Model Details

Model Name: Maximus

Model Type: Transformer-based leveraging Microsoft Phi 3b 128k tokens

Publisher: Awels Engineering

License: MIT

Model Description: Maximus is a sophisticated model designed to help as an AI agent focusing on Maximo Application Suite. It leverages advanced machine learning techniques to provide efficient and accurate solutions. It has been trained on the full docments corpus of MAS 8.5.

Dataset

Dataset Name: awels/maximo_admin_dataset

Dataset Source: Hugging Face Datasets

Dataset License: MIT

Dataset Description: The dataset used to train Maximus consists of all the public documents available on Maximo application suite. This dataset is curated to ensure a comprehensive representation of typical administrative scenarios encountered in Maximo.

Training Details

Training Data: The training data includes 67,000 Questions and Answers generated by the Bonito LLM. The dataset is split into 3 sets of data (training, test and validation) to ensure robust model performance.

Training Procedure: Maximus was trained using supervised learning with cross-entropy loss and the Adam optimizer. The training involved 1 epoch, a batch size of 4, a learning rate of 5.0e-06, and a cosine learning rate scheduler with gradient checkpointing for memory efficiency.

Hardware: The model was trained on a single NVIDIA RTX 4090 graphic card.

Framework: The training was conducted using PyTorch.

Evaluation

Evaluation Metrics: Maximus was evaluated on the training dataset:

epoch = 1.0 total_flos = 64046138GF train_loss = 2.8079 train_runtime = 0:37:48.33 train_samples_per_second = 21.066 train_steps_per_second = 5.267

Performance: The model achieved the following results on the evaluation dataset:

epoch = 1.0 eval_loss = 2.288 eval_runtime = 0:02:05.48 eval_samples = 10773 eval_samples_per_second = 95.338 eval_steps_per_second = 23.836

Intended Use

Primary Use Case: Maximus is intended to be used locally in an agent swarm to colleborate together to solve Maximo Application Suite related problems.

Limitations: While Maximus is highly effective, it may have limitations due to the model size. An 8b model based on Llama 3 is used internally at Awels Engineering.

Downloads last month: 24

GGUF

Model size

4B params

Architecture

phi3

Hardware compatibility

We're not able to determine the quantization variants.

View all variants

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for awels/maximusLLM-3b-128k-gguf

Base model

microsoft/Phi-3-mini-128k-instruct

Quantized

(86)

this model

awels
/

maximusLLM-3b-128k-gguf