Instructions to use sarath-om/sarath_om with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use sarath-om/sarath_om 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 sarath-om/sarath_om
# Run inference directly in the terminal:
llama cli -hf sarath-om/sarath_om

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf sarath-om/sarath_om
# Run inference directly in the terminal:
llama cli -hf sarath-om/sarath_om

Build from source code

git clone https://github.com/ggml-org/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-app
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama serve -hf sarath-om/sarath_om
# Run inference directly in the terminal:
./build/bin/llama cli -hf sarath-om/sarath_om

Use Docker

docker model run hf.co/sarath-om/sarath_om

Llama new
LM Studio
Jan
Ollama
How to use sarath-om/sarath_om with Ollama:
```
ollama run hf.co/sarath-om/sarath_om
```
Unsloth Desktop

How to use sarath-om/sarath_om with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf sarath-om/sarath_om

Configure the model in Pi

# Install Pi:
npm install -g @earendil-works/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "sarath-om/sarath_om"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use sarath-om/sarath_om with Docker Model Runner:
```
docker model run hf.co/sarath-om/sarath_om
```

Lemonade

How to use sarath-om/sarath_om with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull sarath-om/sarath_om

Run and chat with the model

lemonade run user.sarath_om-{{QUANT_TAG}}

List all available models

lemonade list

Hermes Agent

How to use sarath-om/sarath_om with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf sarath-om/sarath_om

Configure Hermes

# Install Hermes:
curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash
hermes setup
# Point Hermes at the local server:
hermes config set model.provider custom
hermes config set model.base_url http://127.0.0.1:8080/v1
hermes config set model.default sarath-om/sarath_om

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use sarath-om/sarath_om with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf sarath-om/sarath_om

Configure OpenClaw

# Install OpenClaw:
npm install -g openclaw@latest
# Register the local server and set it as the default model:
openclaw onboard --non-interactive --mode local \
  --auth-choice custom-api-key \
  --custom-base-url http://127.0.0.1:8080/v1 \
  --custom-model-id "sarath-om/sarath_om" \
  --custom-provider-id llama-cpp \
  --custom-compatibility openai \
  --custom-text-input \
  --accept-risk \
  --skip-health

Run OpenClaw

openclaw agent --local --agent main --message "Hello from Hugging Face"

⚡ Sarath OM

Sarath OM AI

Intelligence. Innovation. Possibility.

An AI project by Sarath OM

Ollama Model · Hugging Face · Official Website · GitHub · LinkedIn

Overview

Sarath OM is an AI project focused on useful, accessible, and extensible AI-powered technology. This repository serves as the Hugging Face entry point for the Sarath OM model project and its associated documentation.

The project emphasizes practical experimentation, developer integration, responsible use, and continuous improvement.

The exact model architecture, underlying weights, supported tasks, training methodology, and measured performance should be determined from the published repository files and verified tests. This card does not assume that the project was trained from scratch or fine-tuned on a particular dataset.

Model Information

Property	Details
Model name	Sarath OM
Repository ID	`sarath-om/sarath-om`
Project	Sarath OM
Developer	Rishabh Srivastava
Architecture	To be verified
Base model	To be verified
Parameter count	To be verified
Model format	Check repository files
Training dataset	Not yet documented
Evaluation results	Not yet documented
License	Confirm the repository license and applicable distribution rights

Project Goals

Sarath OM aims to support the development and evaluation of practical AI-powered experiences.

Potential areas to evaluate, depending on the implementation, include:

AI-assisted applications and prototypes.
Natural-language interaction and information processing.
Developer workflows and software integrations.
Research, experimentation, and educational projects.
Local or hosted inference through supported runtimes.

These are potential use cases, not a guarantee that every listed capability is supported by the current model.

Intended Use

Use the model only for tasks supported by its actual implementation. Before adopting it in an application, test the model with representative prompts and inputs, review the quality of its outputs, and establish appropriate safeguards.

For usage through Ollama, see the Sarath OM Ollama page.

Technical Specifications

Complete these fields using the files published in the repository:

Architecture: Model family and implementation.
Base model: Original model and revision, if applicable.
Weights: Available weight files and formats.
Inputs and outputs: Supported input types and response formats.
Context length: Verified maximum supported context.
Inference framework: Supported runtime and dependencies.
Hardware requirements: Memory and compute requirements based on testing.
Quantization: Available quantization formats, where applicable.

If this repository provides an application or integration that calls an external model rather than distributing its own trained weights, document that distinction clearly.

Installation and Usage

For the Ollama distribution, follow the instructions on the Sarath OM model page.

ollama pull sarath-om/sarath-om
ollama run sarath-om/sarath-om

These commands assume that the requested model identifier is published and available under that exact name. If Ollama reports that the model cannot be found, confirm the model page and identifier.

Do not assume that the model can be loaded directly using Transformers or another Python library until the repository's files and format have been verified.

Training and Data

Training and dataset details are not yet documented in this card.

When confirmed, describe the base model, training or fine-tuning approach, dataset sources, preprocessing, compute resources, and relevant data licenses. Include privacy and data-provenance information where applicable.

Do not publish unsupported claims about training data or model creation.

Evaluation

No verified benchmark results are included in this draft.

A reproducible evaluation should document:

Test datasets and representative tasks.
Metrics and evaluation methodology.
Baseline models and comparison conditions.
Software, hardware, and model configuration.
Observed strengths, errors, and known limitations.

Distinguish measured results from estimates or intended goals.

Limitations and Risks

Model outputs may be incorrect, incomplete, or inconsistent.
Actual capabilities depend on the published implementation and configuration.
Performance may vary with prompts, inputs, and runtime settings.
Bias, privacy, reliability, and misuse risks should be evaluated before deployment.
The model should not be the sole basis for medical, legal, financial, or other high-impact decisions.

Responsible Use

Evaluate Sarath OM for the intended application before production use. Protect confidential information, respect applicable law and third-party rights, and apply appropriate human oversight in consequential workflows.

Review the privacy and retention practices of any external service used with the model before submitting sensitive information.

License

Check the repository's actual license file and metadata before redistribution or commercial use. Verify that all included code, weights, datasets, and third-party components may be distributed under the stated terms. No license is asserted by this draft.

Developer and Official Links

Developer and project owner: Rishabh Srivastava

Official website: https://sarath.openflip.in/
Developer portfolio: https://rishabhsrivastava.figma.site/
Hugging Face profile: https://huggingface.co/sarath-om
GitHub organization: https://github.com/Sarath-Om
LinkedIn: https://www.linkedin.com/company/sarath-ai
Ollama model: https://ollama.com/sarath-om/sarath-om

Versioning

Model name: Sarath OM

Document the release version, source revision, dependencies, compatibility requirements, and changes as they are verified. Keep this card synchronized with the actual model artifacts.

SARATH OM
Intelligence. Innovation. Possibility.

Downloads last month: 286

GGUF

Model size

3B params

Architecture

llama

Hardware compatibility

We're not able to determine the quantization variants.

View all variants

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support