Update README.md

README.md

@@ -1,37 +1,53 @@
 ---
 title: README
-emoji: 📈
+emoji: ⚡
 colorFrom: indigo
 colorTo: pink
 sdk: static
 pinned: false
 ---
 
-# WELCOME!
-> Chase the SOTA pipeline, not the MMLU slop.
-> 
-Meet i3, a state-of-the-are AI model that can be trained in a few hours on a NVIDIA Quadro P100 at the level of LLMs that need days of running on giant GPU farms.
+# Welcome to i3-lab
 
-## Links
-- Repo to the current open-sourced code of i3: https://github.com/FlameF0X/open-i3
-- Discord server: https://discord.gg/qtXApjpaJF
+**"Chase the SOTA pipeline, not the MMLU slop."**
 
-# TODO
-1. [ ] Train `i3-lab/i3-Ethan-it`
-3. [ ] Train `i3-lab/i3-1B`
-2. [ ] Train `i3-lab/i3-7B-A1.6B`
+i3-lab is dedicated to extreme efficiency in LLM architecture. We develop the **i3** model family—state-of-the-art architectures designed to reach high performance levels in hours on consumer-grade hardware (like the NVIDIA Quadro P100) that typically require days on massive GPU clusters.
 
-# If you
-If you end up using our pre-training code, you are legally required to link the repo. The code is licensed under Apache 2.0, specifically under **Section 4(b)** and **4(d)**:
+---
+
+## i3: High-Efficiency Training
+We specialize in hybrid architectures, specifically **RWKV-Attention**, to bypass the quadratic scaling bottlenecks of traditional Transformers.
 
-> (b) You must cause any modified files to carry prominent notices stating that You changed the files;
+* **Fast Iteration:** Trainable in hours, not weeks.
+* **Accessible SOTA:** High performance on legacy/mid-range hardware.
+* **Open Research:** Push the boundaries of what is possible with limited compute.
 
-And most importantly, per **Section 4(d)**:
-> If the Work includes a 'NOTICE' text file... then any Derivative Works that You distribute must include a readable copy of the attribution notices contained within such NOTICE file.
+### Quick Links
+* **Source Code:** [FlameF0X/open-i3](https://github.com/FlameF0X/open-i3)
+* **Community:** [Join our Discord](https://discord.gg/qtXApjpaJF)
+
+---
 
-**This means if you use our architecture (the RWKV-Attention) or weights, you MUST include the attribution link found in the [open-i3](https://github.com/FlameF0X/open-i3) NOTICE file in your documentation or model card.**
+## Roadmap / TODO
+We are currently scaling our architecture through the following milestones:
 
+- [ ] **i3-Ethan-it** — Specialized instruction-tuned variant.
+- [ ] **i3-1B** — Our first major scale-up.
+- [ ] **i3-7B-A1.6B** — Mixture of Experts / Sparsity testing.
 
 ---
 
-Made with ❤ and DETERMINATION by Daniel.
+## Usage & Attribution
+The `open-i3` codebase is licensed under **Apache 2.0**. We believe in open-source, but we value attribution. 
+
+If you use our architecture (RWKV-Attention) or our weights, you are required per **Section 4(b)** and **4(d)** to:
+1.  Carry prominent notices of any modifications.
+2.  Include a readable copy of the attribution notices from our **NOTICE** file.
+
+> [!IMPORTANT]
+> You **must** include the attribution link found in the [open-i3 GitHub](https://github.com/FlameF0X/open-i3) in your documentation or model card.
+
+---
+<p align="center">
+  Made with ❤️ and <b>DETERMINATION</b> by Daniel.
+</p>