Spaces:
Sleeping
Sleeping
| title: AxionX Digital β Model Training & Evaluation Suite | |
| emoji: π | |
| colorFrom: blue | |
| colorTo: purple | |
| sdk: streamlit | |
| sdk_version: "1.36.0" | |
| app_file: streamlit_hub.py | |
| pinned: true | |
| license: mit | |
| short_description: Fine-tuning and evaluation demos for domain-specific LLMs | |
| tags: | |
| - streamlit | |
| - huggingface | |
| - fine-tuning | |
| - peft | |
| - lora | |
| - evaluation | |
| - llm | |
| - axionx | |
| π AxionX Digital β Model Training & Evaluation Suite | |
| A live collection of open-source LLM fine-tuning and evaluation demos β all running directly on Hugging Face Spaces. | |
| π‘ Included Demos | |
| Project Description Tech | |
| π° FinanceGPT Fine-tuned FLAN-T5 for financial Q&A and summarization LoRA Β· HF Transformers | |
| βοΈ LegalDoc Summarizer Clause-level summarization using the CUAD dataset FLAN-T5 Β· PEFT | |
| ποΈ RetailGPT Evaluator Benchmarking retail-QA models + leaderboard UI Evaluation Β· Streamlit | |
| π€ AxionX Gradio Demo Interactive Gradio showcase of AxionXβs capabilities Gradio Β· Transformers | |
| π Live App | |
| π’ Launch here: | |
| π https://huggingface.co/spaces/deepsodha/axionx-demo | |
| Once open, use the sidebar to switch between: | |
| π° FinanceGPT | |
| βοΈ LegalDoc Summarizer | |
| ποΈ RetailGPT Evaluator | |
| π€ AxionX Gradio Demo | |
| Each section runs instantly inside the browser β no local setup required. | |
| π§© Project Structure | |
| vbnet | |
| Copy code | |
| axionx-demo/ | |
| βββ financegpt/ β Finance fine-tuning demo | |
| βββ legaldoc_summarizer/ β Legal text summarization demo | |
| βββ retailgpt_evaluator/ β Retail chatbot evaluator | |
| βββ shared/ β Common helpers & requirements | |
| βββ streamlit_hub.py β Unified Streamlit dashboard | |
| βββ app.py β Original AxionX Gradio demo | |
| π’ About AxionX Digital | |
| AxionX Digital builds fine-tuned AI models and evaluation dashboards for enterprises in Finance, Legal, and Retail. | |
| We specialize in: | |
| π§ Domain-specific LLM fine-tuning | |
| βοΈ LoRA / PEFT optimization | |
| π Evaluation & leaderboard dashboards | |
| π Lightweight Hugging Face deployments | |
| π Website: https://axionxdigital.com | |
| π§ Contact: [email protected] | |
| π§ Tech Stack | |
| Transformers Β· PEFT (LoRA) Β· Streamlit Β· Gradio Β· Hugging Face Datasets Β· Python 3.10+ | |
| πͺͺ License | |
| Released under the MIT License β feel free to fork and adapt with attribution. | |
| β¨ Β© 2025 AxionX Digital β Innovating Tomorrow. |