Instructions to use NithinAI12/Nithin-ASI-v1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- llama.cpp
How to use NithinAI12/Nithin-ASI-v1 with llama.cpp:
Install (macOS, Linux)
curl -LsSf https://llama.app/install.sh | sh # Start a local OpenAI-compatible server with a web UI: llama serve -hf NithinAI12/Nithin-ASI-v1 # Run inference directly in the terminal: llama cli -hf NithinAI12/Nithin-ASI-v1
Install from WinGet (Windows)
winget install llama.cpp # Start a local OpenAI-compatible server with a web UI: llama serve -hf NithinAI12/Nithin-ASI-v1 # Run inference directly in the terminal: llama cli -hf NithinAI12/Nithin-ASI-v1
Use pre-built binary
# Download pre-built binary from: # https://github.com/ggerganov/llama.cpp/releases # Start a local OpenAI-compatible server with a web UI: ./llama-server -hf NithinAI12/Nithin-ASI-v1 # Run inference directly in the terminal: ./llama-cli -hf NithinAI12/Nithin-ASI-v1
Build from source code
git clone https://github.com/ggerganov/llama.cpp.git cd llama.cpp cmake -B build cmake --build build -j --target llama-server llama-cli # Start a local OpenAI-compatible server with a web UI: ./build/bin/llama-server -hf NithinAI12/Nithin-ASI-v1 # Run inference directly in the terminal: ./build/bin/llama-cli -hf NithinAI12/Nithin-ASI-v1
Use Docker
docker model run hf.co/NithinAI12/Nithin-ASI-v1
- LM Studio
- Jan
- Ollama
How to use NithinAI12/Nithin-ASI-v1 with Ollama:
ollama run hf.co/NithinAI12/Nithin-ASI-v1
- Unsloth Desktop
- Docker Model Runner
How to use NithinAI12/Nithin-ASI-v1 with Docker Model Runner:
docker model run hf.co/NithinAI12/Nithin-ASI-v1
- Lemonade
How to use NithinAI12/Nithin-ASI-v1 with Lemonade:
Pull the model
# Download Lemonade from https://lemonade-server.ai/ lemonade pull NithinAI12/Nithin-ASI-v1
Run and chat with the model
lemonade run user.Nithin-ASI-v1-{{QUANT_TAG}}List all available models
lemonade list
- Atomic Chat
YAML Metadata Warning:empty or missing yaml metadata in repo card
Check out the documentation for more information.
π§ Nithin-ASI (Artificial Super Intelligence)
Standalone, Fully Localized Enterprise Reasoning Models via GGUF & PyTorch.
Engineered for zero-latency execution, zero recurring API costs, and absolute data privacy.
π The Vision
The current AI ecosystem relies heavily on cloud-hosted API wrappers, creating high recurring costs and critical data privacy risks. Nithin-ASI was engineered as a decentralized alternative. By fine-tuning large language models and compiling them into highly compressed GGUF formats, Nithin-ASI enables complex neural reasoning directly on edge hardware without internet dependency.
βοΈ Core Architecture Matrix
graph TD;
A[Raw Datasets & Logic] -->|PyTorch Fine-Tuning| B(Nithin-ASI Base Weights);
B -->|Quantization Engine| C{GGUF Format};
C -->|Deployment| D[Local Hardware / Edge Node];
C -->|Deployment| E[Nithin-Studio Web App];
D --> F[Zero-Latency Execution];
E --> F;
style C fill:#00f2fe,stroke:#fff,stroke-width:2px,color:#000
style F fill:#32d74b,stroke:#fff,stroke-width:2px,color:#000
Key Technical InnovationsGGUF Compression: Advanced quantization algorithms reduce VRAM overhead by 70%, allowing enterprise-grade reasoning on standard consumer hardware.Air-Gapped Privacy: Models run 100% locally. Zero telemetry data is transmitted externally.Hardware-Agnostic: Runs on CPU-only edge devices or accelerates dynamically via Apple Metal, CUDA, or ROCm.π Performance Benchmarks (Local Execution)MetricCloud API (Standard)Nithin-ASI (Local Edge)AdvantageData PrivacySent to Cloud Server100% Kept on DeviceSecureRecurring Cost$0.03 / 1k Tokens$0.00 (Free)Infinite ROINetwork Latency~200ms - 800ms0ms (Air-Gapped)Instantπ» Quick Start: Booting Nithin-ASIDeploy the weights using standard llama.cpp terminal commands:Bash# Boot the model into an interactive terminal session
./main -m ./models/Nithin_ASI_Final.gguf -n 256 --repeat_penalty 1.1 --color -i -r "User:"
"Work Smart, Not Hard. Leverage open-weight structures to build customized, localized reasoning engines."
β Engineered by G.V. Nithin
- Downloads last month
- 70
We're not able to determine the quantization variants.