Deploy official Experimentlab organization profile card
Browse files
README.md
ADDED
|
@@ -0,0 +1,82 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
title: Experimentlab
|
| 3 |
+
emoji: ⚛️
|
| 4 |
+
colorFrom: indigo
|
| 5 |
+
colorTo: purple
|
| 6 |
+
sdk: static
|
| 7 |
+
pinned: false
|
| 8 |
+
---
|
| 9 |
+
|
| 10 |
+
<div align="center">
|
| 11 |
+
|
| 12 |
+
# ⚛️ Experimentlab (Experimentlab.in)
|
| 13 |
+
### Sovereign Artificial Intelligence for Science, Technology, Engineering & Mathematics (STEM)
|
| 14 |
+
|
| 15 |
+
[](https://experimentlab.in)
|
| 16 |
+
[](https://github.com/shreyansh001boy-tech/vigyan_ai)
|
| 17 |
+
[](https://experimentlab.in/desktop)
|
| 18 |
+
|
| 19 |
+
---
|
| 20 |
+
|
| 21 |
+
</div>
|
| 22 |
+
|
| 23 |
+
## 🌌 Mission Statement
|
| 24 |
+
|
| 25 |
+
**Experimentlab** is dedicated to building sovereign, open-weights foundation models, efficient Small Language Models (SLMs), and offline workstations designed specifically for **STEM researchers, engineers, mathematicians, and students**.
|
| 26 |
+
|
| 27 |
+
Our core philosophy bridges deep academic theory with extreme computational efficiency:
|
| 28 |
+
* **Zero Cloud Dependence**: 100% offline, privacy-first execution on consumer laptops and edge devices.
|
| 29 |
+
* **Mathematical Exactitude**: Deterministic CAS (Computer Algebra System) verification for symbolic calculus, linear algebra, and discrete math.
|
| 30 |
+
* **Sovereign Efficiency**: Pioneer Sparse Mixture-of-Experts (MoE) and 1.58-bit ternary {-1, 0, +1} BitNet quantization running natively on CPUs and NPUs.
|
| 31 |
+
|
| 32 |
+
---
|
| 33 |
+
|
| 34 |
+
## 🏛️ Flagship Model Architectures
|
| 35 |
+
|
| 36 |
+
```mermaid
|
| 37 |
+
graph LR
|
| 38 |
+
A["Raw STEM Corpora (Ai2 Dolma, SymPy CAS, arXiv)"] --> B["Vigyan Sovereign Pretraining"]
|
| 39 |
+
B --> C["Vigyan MoE (1.5B - 3B 4x Experts)"]
|
| 40 |
+
B --> D["BitNet 1.58-Bit Ternary QAT"]
|
| 41 |
+
B --> E["Doctoral STEM Series (7B - 32B)"]
|
| 42 |
+
C --> F["Edge CPU & NPU (GGUF / bitnet.cpp)"]
|
| 43 |
+
D --> F
|
| 44 |
+
E --> G["Vigyan Pro Workstation"]
|
| 45 |
+
```
|
| 46 |
+
|
| 47 |
+
### 1. Sparse Mixture of Experts (MoE)
|
| 48 |
+
* **[`Vigyan-1.5B-4x-MoE`](https://huggingface.co/Experimentlab/Vigyan-1.5B-4x-MoE)**: Ultra-fast 4-expert routed SLM delivering 7B-class STEM reasoning at a fraction of inference memory.
|
| 49 |
+
* **[`Vigyan-3B-4x-MoE`](https://huggingface.co/Experimentlab/Vigyan-3B-4x-MoE)**: Advanced multi-expert router tuned on symbolic logic and scientific problem-solving.
|
| 50 |
+
|
| 51 |
+
### 2. 1.58-Bit Ternary BitNet Architectures
|
| 52 |
+
* **[`vigyan-3b-1.58bit-qat`](https://huggingface.co/Experimentlab/vigyan-3b-1.58bit-qat)**: Pure 1.58-bit weights {-1, 0, +1} trained via Quantization-Aware Training (QAT). Eliminates power-hungry matrix multiplication in favor of integer addition, enabling offline execution on low-power ARM and x86 chips.
|
| 53 |
+
|
| 54 |
+
### 3. Doctoral & Silicon Hardware Series
|
| 55 |
+
* **[`olmo2-7b-phd-pure-math`](https://huggingface.co/Experimentlab/olmo2-7b-phd-pure-math)**: Specialization in pure mathematics, formal proofs, topology, and abstract algebra.
|
| 56 |
+
* **[`olmo2-7b-silicon-rtl-eda`](https://huggingface.co/Experimentlab/olmo2-7b-silicon-rtl-eda)**: Hardware silicon EDA specialist for Verilog RTL synthesis, timing constraints, and digital logic verification.
|
| 57 |
+
* **[`olmo2-7b-biomed-chem`](https://huggingface.co/Experimentlab/olmo2-7b-biomed-chem)**: Deep biochemistry, computational medicine, and molecular modeling.
|
| 58 |
+
|
| 59 |
+
---
|
| 60 |
+
|
| 61 |
+
## 💻 Vigyan Pro Desktop: Offline AI Workstation
|
| 62 |
+
|
| 63 |
+
We build **Vigyan Pro Desktop**, a sovereign offline AI workstation featuring:
|
| 64 |
+
* **Interactive WaveDrom Digital Timing Studio** for electrical engineering and hardware timing diagrams.
|
| 65 |
+
* **Native SymPy CAS Mathematical Engine** for symbolic calculus, derivatives, integrals, and LaTeX rendering.
|
| 66 |
+
* **Hybrid RAG & Research Library** for instant citation-backed analysis across local academic PDFs.
|
| 67 |
+
|
| 68 |
+
Download and install the Community Edition directly from [experimentlab.in/desktop](https://experimentlab.in/desktop).
|
| 69 |
+
|
| 70 |
+
---
|
| 71 |
+
|
| 72 |
+
## 📂 Curated Collections
|
| 73 |
+
Explore our verified models grouped by domain:
|
| 74 |
+
1. **[🌟 Sovereign STEM SLMs & MoE Architectures](https://huggingface.co/collections/Experimentlab)**
|
| 75 |
+
2. **[🔬 Doctoral Mathematics & Silicon EDA](https://huggingface.co/collections/Experimentlab)**
|
| 76 |
+
3. **[📚 Open Datasets & Verified Research Whitepapers](https://huggingface.co/collections/Experimentlab)**
|
| 77 |
+
|
| 78 |
+
---
|
| 79 |
+
|
| 80 |
+
<div align="center">
|
| 81 |
+
<sub>Built with passion for sovereign open-source STEM research by Experimentlab.in • All models strictly open-weights.</sub>
|
| 82 |
+
</div>
|