Indic LM Platform

In plain words

Ambari is a bilingual Kannada-English large language model from CognitiveLab, built to bridge the gap for localised state-language adaptation.

How to use it

Before you start

Prerequisites required to run the model locally

To run Ambari locally on your computer, the following requirements must be met:

Install and set up Ollama.
Ensure your computer has enough RAM to run the model you plan to download.

For 3B models: Minimum 8 GB of RAM.
For 7B–8B models: Minimum 16 GB of RAM.
For 13B–32B models: Minimum 32 GB of RAM.

Quantised model: It is recommended to use 4-bit Quantized, K-measurement, Small model format. Easy to download and to run your system.
Storage: 5 to 20 GB of free SSD space to store downloaded model files.
Interface: Terminal or command line utility to run basic startup prompts.

No technical setup needed — anyone can try Ambari using below mentioned steps.

Downloading the Ambari GGUF file

Go to the Ambari GGUF file on Hugging Face. Click the copy icon next to the repository name to copy the model path. (See image below.)

Run the model in your terminal

Open a terminal window and run the following command and paste the copied GGUF repository as mentioned below:

ollama run hf.co/PASTE_THE_COPIED_HERE

Press Enter.

Note: After pasting the repository, your command will look like this:

ollama run hf.co/tensorblock/Ambari-7B-Instruct-v0.1-sharded-GGUF

Wait for the model to load

Ollama will download and initialise the model. Once ready, you'll see the following prompt:

>>> Send a message (/? for help)

Start chatting

Type your message at the prompt and press Enter to interact with the model.

Exit the chat

To end the session, type:

/bye

Then press Enter.

Languages & scripts supported

Works in these languages, in both native script and Roman typing.

ಕನ್ನಡ KannadaEnglish English+ Roman & code-mixed

Strengths & limits

An honest look at what it does well and where it struggles.

Good at

✓Native language processing

✓Text generation

Where it struggles

!Highly complex code generation

!Low-resource dialects without fine-tuning

Technical details & licence

Training dataPretrained on ~500M new Kannada tokens added to Llama 2, followed by bilingual next-token-prediction training and instruction fine-tuning with DPO on Wikipedia, Alpaca, Dolly, and a Kannada-translated HH-RLHF dataset.

SizeBuilt on Meta's Llama 2 7B architecture with a custom Kannada SentencePiece tokenizer (vocabulary expanded to ~49,600 tokens), released as base and DPO instruction-tuned variants.

LicenceInherits Meta's Llama 2 licence as a derivative model — not a fully permissive licence; commercial use is subject to Llama 2's terms and acceptable-use restrictions.

Commercial costFree to self-host; hosted demo available

Version historyv0.1 (base) · Jan 2024 — First bilingual Kannada-English base model releasedv0.2 (instruct) · March 2024 — Instruction-tuned with DPO

About the maker

Who builds and maintains Ambari.

CognitiveLab

2 models

CognitiveLab is a Bengaluru-based, open-source-first AI research lab founded by Aditya Kolavi. It is best known for Ambari, one of the first bilingual Kannada-English language models (built on Llama-2), and for open tools like OmniParse and the multilingual OCR project Nayana — the latter winning a 2024 Meta Llama Impact Grant, announced in April 2025.

View their models →See all organisations →