# š¤ Hi,
I build tiny, hyper-optimized, and privacy-first local Small Language Model (SLM) agents designed to run offline on everyday CPU hardware.
### š ļø What I'm Building
* **[SLM Agents](https://huggingface.co/spaces/spcv/slm-agents)**: A monorepo ecosystem of 26 local CPU-optimized agents (Text-to-SQL, RAG, Web Scrapers, Voice Assistants, and CLI agents) running offline using ONNX Runtime GenAI.
* **Fine-Tuned SLMs**: Training and exporting specialized, lightweight models (like [Qwen-2.5-Coder-Text2SQL ONNX](https://huggingface.co/spcv/qwen2.5_coder_text2sql_onnx)) for high-performance localized edge computing.
### ā” Technical Focus
`Small Language Models (SLMs)` ⢠`ONNX Runtime GenAI` ⢠`Offline Inference` ⢠`Local RAG` ⢠`CPU Optimization` ⢠`Agentic Workflows`
---
š¬ *Let's build the future of decentralized, offline-first AI together! Star the space and check out the code!*