Triangle104/Ava-1.0-8B-Q4_K_S-GGUF

This model was converted to GGUF format from Spestly/Ava-1.0-8B using llama.cpp via the ggml.ai's GGUF-my-repo space. Refer to the original model card for more details on the model.

---

Model details:

Ava 1.0

Ava 1.0 is an advanced AI model fine-tuned on the Mistral architecture, featuring 8 billion parameters. Designed to be smarter, stronger, and swifter, Ava 1.0 excels in tasks requiring comprehension, reasoning, and language generation, making it a versatile solution for various applications.

Key Features

Compact Yet Powerful:
    With 8 billion parameters, Ava 1.0 strikes a balance between computational efficiency and performance.

Enhanced Reasoning Capabilities:
    Fine-tuned to provide better logical deductions and insightful responses across multiple domains.

Optimized for Efficiency:
    Faster inference and reduced resource requirements compared to larger models.

Use Cases

Conversational AI: Natural and context-aware dialogue generation.
Content Creation: Generate articles, summaries, and creative writing.
Educational Tools: Assist with problem-solving and explanations.
Data Analysis: Derive insights from structured and unstructured data.

Technical Specifications

Model Architecture: Ministral-8B-Instruct-2410
Parameter Count: 8 Billion
Training Dataset: A curated dataset spanning diverse fields, including literature, science, technology, and general knowledge.
Framework: Hugging Face Transformers

Usage

To use Ava 1.0, integrate it into your Python environment with Hugging Face's transformers library:

Use a pipeline as a high-level helper

from transformers import pipeline

messages = [ {"role": "user", "content": "Who are you?"}, ] pipe = pipeline("text-generation", model="Spestly/Ava-1.0-8B") pipe(messages)

Load model directly

from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Spestly/Ava-1.0-8B") model = AutoModelForCausalLM.from_pretrained("Spestly/Ava-1.0-8B")

Future Plans

Continued optimization for domain-specific applications.
Expanding the model's adaptability and generalization capabilities.

Contributing

We welcome contributions and feedback to improve Ava 1.0. If you'd like to get involved, please reach out or submit a pull request.

License

This model is licensed under Mistral Research License. Please review the license terms before usage.

---

Use with llama.cpp

Install llama.cpp through brew (works on Mac and Linux)

brew install llama.cpp

Invoke the llama.cpp server or the CLI.

CLI:

llama-cli --hf-repo Triangle104/Ava-1.0-8B-Q4_K_S-GGUF --hf-file ava-1.0-8b-q4_k_s.gguf -p "The meaning to life and the universe is"

Server:

llama-server --hf-repo Triangle104/Ava-1.0-8B-Q4_K_S-GGUF --hf-file ava-1.0-8b-q4_k_s.gguf -c 2048

Note: You can also use this checkpoint directly through the usage steps listed in the Llama.cpp repo as well.

Step 1: Clone llama.cpp from GitHub.

git clone https://github.com/ggerganov/llama.cpp

Step 2: Move into the llama.cpp folder and build it with LLAMA_CURL=1 flag along with other hardware-specific flags (for ex: LLAMA_CUDA=1 for Nvidia GPUs on Linux).

cd llama.cpp && LLAMA_CURL=1 make

Step 3: Run inference through the main binary.

./llama-cli --hf-repo Triangle104/Ava-1.0-8B-Q4_K_S-GGUF --hf-file ava-1.0-8b-q4_k_s.gguf -p "The meaning to life and the universe is"

or

./llama-server --hf-repo Triangle104/Ava-1.0-8B-Q4_K_S-GGUF --hf-file ava-1.0-8b-q4_k_s.gguf -c 2048