Update README.md

README.md

@@ -9,116 +9,140 @@ pinned: false
 
 <div align="center">
 
-<br>
-
-<img src="logo.png" alt="RuVI Labs logo" width="120">
+<img src="logo.png" alt="RuVI Labs logo" width="96">
 
-<br><br>
+<br>
 
 <img src="Ruvi.png" alt="RuVI Labs" width="100%">
 
-<br>
+# RuVI Labs
 
-**B U I L D &nbsp; G O O D &nbsp; M O D E L S &nbsp; F O R &nbsp; P E O P L E**
-
-<br>
+**Build good models for people.**
 
+<sub>A research lab from India · language models, built the honest way</sub>
 
+<br>
 
-![India](https://img.shields.io/badge/MADE_IN-INDIA-000000?style=flat-square&labelColor=000000)
+![India](https://img.shields.io/badge/MADE_IN-INDIA-000000?style=for-the-badge&labelColor=000000)
+![Weights](https://img.shields.io/badge/WEIGHTS-OPEN-ffffff?style=for-the-badge&labelColor=000000)
+![Data](https://img.shields.io/badge/DATA-OPEN-000000?style=for-the-badge&labelColor=000000)
+![Scratch](https://img.shields.io/badge/BUILT-FROM_SCRATCH-ffffff?style=for-the-badge&labelColor=000000)
 
+</div>
 
+---
 
+## 01 · Who we are
 
-![Weights](https://img.shields.io/badge/WEIGHTS-OPEN-ffffff?style=flat-square&labelColor=000000)
+RuVI Labs is a small, independent research lab based in India. We design, train and release language models end to end: architecture, data, training and inference, with no borrowed foundations.
 
+We build for the people who will actually use the models: students, developers, and teams working with limited compute. If a model can't run where its users are, it isn't finished.
 
+<div align="center">
 
+*Small lab, serious work.*
 
-![Data](https://img.shields.io/badge/DATA-OPEN-000000?style=flat-square&labelColor=000000)
+</div>
 
+---
 
+## 02 · Pragya
 
+<div align="center">
 
-![Scratch](https://img.shields.io/badge/BUILT-FROM_SCRATCH-ffffff?style=flat-square&labelColor=000000)
+# P R A G Y A
 
+**A family of language models**
+**trained end to end on open data**
 
+</div>
 
-<br>
+<!-- TODO: replace the placeholders below with real sizes, context lengths and links as they ship -->
 
-</div>
+| Model | Status | Notes |
+|:--|:-:|:--|
+| **Pragya** (flagship) | 🔨 Training | Open weights, open data |
+| **Pragya Small** | 🧪 Planned | Built for low-resource hardware |
+| **Pragya Instruct** | 🧪 Planned | Chat and instruction tuned |
 
-<br>
+<sub>Status key: ✅ Released · 🔨 Training · 🧪 Planned</sub>
 
-```
-━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
-```
+---
 
-<div align="center">
+## 03 · Our principles
 
-### &nbsp;
+<table>
+<tr>
+<td width="33%" align="center">
 
-*A research lab from India.*
-*Language models, built the honest way,*
-*from scratch, for the people who will actually use them.*
+**◼ Open weights**
 
-### &nbsp;
+Free to use, free to build on.
 
-</div>
+</td>
+<td width="33%" align="center">
 
-```
-━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
-```
+**◻ Open data**
 
-<br>
+Nothing hidden in the corpus.
 
-## 01 &nbsp;/&nbsp; WHAT WE BUILD
+</td>
+<td width="33%" align="center">
 
-<br>
+**◼ From scratch**
 
-<div align="center">
+No borrowed foundations.
 
-# P R A G Y A
+</td>
+</tr>
+</table>
 
-**a family of language models**
-**trained end-to-end on open data**
+### What "from scratch" means to us
 
-</div>
+- **Own architecture.** Designed and implemented in-house, not a fine-tune of someone else's checkpoint.
+- **Own training run.** Pretrained by us on data we can document.
+- **Own tooling.** Training, evaluation and inference code built and released alongside the weights.
 
-<br>
+### What "open" means to us
 
-| | | |
-|:-:|:-:|:-:|
-| **◼ OPEN WEIGHTS** | **◻ OPEN DATA** | **◼ FROM SCRATCH** |
-| Free to use,<br>free to build on | Nothing hidden<br>in the corpus | No borrowed<br>foundations |
+- Weights you can download, run and modify.
+- A data recipe you can inspect and question.
+- Honest reporting, including what didn't work.
 
-<br>
+---
 
-> *Just work that stands on its own.*
+## 04 · What we believe
 
-<br>
+> **Good models are *different*.**
 
-## 02 &nbsp;/&nbsp; WHAT WE BELIEVE
+**I.** Trained to perform differently.
+**II.** Built with the motive to undo what we assume is possible in AI.
+**III.** Made to achieve, and made to perform.
 
-<br>
+---
 
-<div align="center">
+## 05 · What we're working on
 
-# Good models are *different.*
+- 🔨 **Pretraining the Pragya family** on open data
+- 🧪 **Efficient architectures** that make small models punch above their size
+- 🧪 **Quantization and fast inference** so models run on modest hardware
+- 🧪 **Open evaluation**: transparent benchmarks and model cards
 
-</div>
+<!-- TODO: edit this list to match your real roadmap -->
 
-<br>
+---
 
-**I.** &nbsp; Trained to perform differently.
-**II.** &nbsp; Built with the motive to undo what we assume is possible in AI.
-**III.** &nbsp; Made to achieve, and made to perform.
+## 06 · Roadmap
 
-<br>
+| Phase | Goal |
+|:--|:--|
+| **Now** | Train and iterate on the first Pragya models |
+| **Next** | Release weights, data recipe and model cards |
+| **Later** | Instruction-tuned and efficient variants |
 
-## 03 &nbsp;/&nbsp; STATUS
+---
 
-<br>
+## 07 · Quick facts
 
 | | |
 |:--|:--|
@@ -129,24 +153,37 @@ pinned: false
 | **License** | Open weights |
 | **Mode** | Building quietly. Shipping publicly. |
 
-<br>
+---
 
-```
-━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
-```
+## 08 · FAQ
 
-<div align="center">
+**Are the models really trained from scratch?**
+Yes. Architecture, pretraining and tooling are our own.
 
-<br>
+**Can I use the weights commercially?**
+Check the license on each model card. Our goal is permissive, open release.
 
-<img src="logo.png" alt="RuVI Labs" width="64">
+**Can I contribute?**
+Yes. Open a discussion on any repo here, or reach out below.
 
-<br><br>
+<!-- TODO: confirm the license answer once licenses are final -->
 
-**R U V I &nbsp; L A B S**
+---
 
-*small lab, serious work.*
+## 09 · Follow along
+
+Follow the organization on Hugging Face to get notified when models and datasets drop.
+
+<!-- TODO: add links, e.g. GitHub · X · email -->
+
+<div align="center">
 
 <br>
 
-</div>
+<img src="logo.png" alt="RuVI Labs" width="56">
+
+**RUVI LABS**
+
+*small lab, serious work.*
+
+</div>