🇮🇳 SaShi 1.0 (8B GGUF) — India's Own Sovereign AI
SaShi 1.0 is an independent, sovereign conversational AI model engineered in Bharat by Two Brothers (named 'Sa-Shi') from Varanasi (Banaras), Uttar Pradesh.
-green?style=for-the-badge){kind=icon}
📌 Origin & Vision
- Origin of Name: Sa-Shi (Named after the two brothers who developed & trained this model).
- Creators & Developers: Built and trained by Two Brothers in Varanasi (Kashi / Banaras), Uttar Pradesh, India.
- Specialization: Real Indian ground realities, fluent conversational Hinglish/Hindi, coding, and culturally rooted guidance.
- Format: GGUF
Q4_K_M(4-bit quantized for ultra-fast inference).
⚙️ Technical Specifications
| Parameter | Value |
|---|---|
| Model Name | SaShi 1.0 |
| Parameters | 8.03 Billion (8B) |
| Format | GGUF Q4_K_M |
| Creators | Two Brothers (Sa-Shi) • Varanasi, UP 🇮🇳 |
| Primary Languages | Hinglish (Roman Hindi), Hindi (हिन्दी), English |
🛠️ Usage with Ollama
- Download the
SaShi-1.0.gguffile. - Create a
Modelfile:
FROM ./SaShi-1.0.gguf
TEMPLATE """{{ if .System }}<|start_header_id|>system<|end_header_id|>
{{ .System }}<|eot_id|>{{ end }}{{ if .Prompt }}<|start_header_id|>user<|end_header_id|>
{{ .Prompt }}<|eot_id|>{{ end }}<|start_header_id|>assistant<|end_header_id|>
"""
PARAMETER stop "<|start_header_id|>"
PARAMETER stop "<|end_header_id|>"
PARAMETER stop "<|eot_id|>"
- Build and run:
ollama create sashi-1.0 -f Modelfile
ollama run sashi-1.0
🐍 Usage with Python (llama-cpp-python)
from llama_cpp import Llama
llm = Llama(
model_path="SaShi-1.0.gguf",
n_ctx=2048,
n_gpu_layers=-1
)
output = llm(
"User: SaShi, tumhare baare me batao.\nSaShi:",
max_tokens=128,
stop=["User:", "<|eot_id|>"]
)
print(output["choices"][0]["text"])
SaShi was born out of a dream to prove that developers from India's cultural capital, Varanasi, can build sovereign, world-class artificial intelligence tailored for the unique spirit and ground realities of Bharat.