commit 43c771d2e3f0013ecac5b14db76168d685d52a60
Author: Inference-X <admin@inference-x.com>
Date:   Wed Feb 25 17:18:43 2026 +0000

    docs: initial commit

diff --git a/README.md b/README.md
new file mode 100644
index 0000000..57425b5
--- /dev/null
+++ b/README.md
@@ -0,0 +1,50 @@
+# IX Tools
+
+**Utilities for the Inference-X ecosystem**
+
+[Inference-X](https://inference-x.com) · [Community](https://git.inference-x.com/inference-x-community)
+
+---
+
+A collection of tools for working with GGUF models, benchmarking hardware, managing Inference-X deployments, and integrating with common workflows.
+
+## Tools
+
+| Tool | Description |
+|---|---|
+| `ix-bench` | Hardware benchmark — measures t/s across backends |
+| `ix-convert` | Convert models to GGUF from SafeTensors, PyTorch |
+| `ix-pull` | Download GGUF models from Hugging Face |
+| `ix-serve` | Production-ready IX server wrapper with auth, logging |
+| `ix-proxy` | Load balancer across multiple IX instances |
+| `ix-monitor` | Dashboard — GPU usage, t/s, active connections |
+| `ix-chat` | Terminal chat UI with history and markdown rendering |
+| `ix-embed` | Batch embedding generation tool |
+
+## Quick Start
+
+```bash
+git clone https://git.inference-x.com/inference-x/ix-tools
+cd ix-tools && make
+
+# Benchmark your hardware
+./ix-bench --all-backends
+
+# Download a model
+./ix-pull qwen2.5-7b-instruct-q4_k_m.gguf
+
+# Start production server
+./ix-serve --model model.gguf --port 8080 --workers 4
+
+# Monitor running instances
+./ix-monitor --host localhost:8080
+```
+
+## Hardware Benchmark Reference (Feb 2026)
+
+Run `./ix-bench` to measure your actual hardware.  
+Community results: [git.inference-x.com/inference-x-community/ix-scout](https://git.inference-x.com/inference-x-community/ix-scout)
+
+---
+
+[inference-x.com](https://inference-x.com) · MIT License