Instructions to use Bonenk/indic-translate-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use Bonenk/indic-translate-GGUF 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 Bonenk/indic-translate-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf Bonenk/indic-translate-GGUF:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf Bonenk/indic-translate-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf Bonenk/indic-translate-GGUF:Q4_K_M

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 Bonenk/indic-translate-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama cli -hf Bonenk/indic-translate-GGUF:Q4_K_M

Use Docker

docker model run hf.co/Bonenk/indic-translate-GGUF:Q4_K_M

Llama new
LM Studio
Jan
Ollama
How to use Bonenk/indic-translate-GGUF with Ollama:
```
ollama run hf.co/Bonenk/indic-translate-GGUF:Q4_K_M
```
Unsloth Desktop

How to use Bonenk/indic-translate-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf Bonenk/indic-translate-GGUF:Q4_K_M

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": "Bonenk/indic-translate-GGUF:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use Bonenk/indic-translate-GGUF with Docker Model Runner:
```
docker model run hf.co/Bonenk/indic-translate-GGUF:Q4_K_M
```

Lemonade

How to use Bonenk/indic-translate-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull Bonenk/indic-translate-GGUF:Q4_K_M

Run and chat with the model

lemonade run user.indic-translate-GGUF-Q4_K_M

List all available models

lemonade list

Hermes Agent

How to use Bonenk/indic-translate-GGUF with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf Bonenk/indic-translate-GGUF:Q4_K_M

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 Bonenk/indic-translate-GGUF:Q4_K_M

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use Bonenk/indic-translate-GGUF with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf Bonenk/indic-translate-GGUF:Q4_K_M

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 "Bonenk/indic-translate-GGUF:Q4_K_M" \
  --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"

Indic Translate — GGUF (llama.cpp)

Quantized bodhan-ai/indic-translate for llama.cpp. Converted from the upstream weights at revision 3c5c3b4e2747b0fca133accf077132326c88b267.

Text only. Upstream is Gemma4ForConditionalGeneration (vision + audio towers); those towers are not part of a translation workload, so the GGUF carries the text decoder only (architecture = gemma4, 684 tensors).
Whole documents work: upstream advertises up to 32k input tokens.

Files

File	Quant	Size	Notes
`indic-translate-Q4_K_M.gguf`	Q4_K_M	5.30 GB	smallest; best speed/quality trade-off
`indic-translate-Q6_K.gguf`	Q6_K	6.17 GB	near-lossless
`indic-translate-Q8_0.gguf`	Q8_0	7.97 GB	effectively lossless
`indic-translate-f16.gguf`	F16	14.93 GB	reference / re-quantize from this

All four were loaded and generated with llama-cli build b11798-9febe63d5.

Usage

llama-cli -m indic-translate-Q4_K_M.gguf \
  -ngl 99 -c 4096 --jinja --temp 0 \
  -p "Translate the following text into Hindi: Good morning, my friend."
# → सुप्रभात, मेरे दोस्त।

Server mode:

llama-server -m indic-translate-Q4_K_M.gguf -ngl 99 -c 8192 --jinja --port 8080

curl http://127.0.0.1:8080/v1/chat/completions -H 'Content-Type: application/json' -d '{
  "messages": [{"role": "user",
                "content": "Translate the following text into Hindi: Good morning, my friend."}],
  "temperature": 0
}'

Prompting

Name the target language in the instruction — upstream exposes a prompt name per language and script (Hindi, Tamil, Bengali, Urdu, Konkani, …):

Translate the following text into Hindi: <source text>

Long inputs are fine; the whole document goes in one request.

License

The upstream Indic Open Model License v1.0 applies — see LICENSE-INDIC-OPEN-MODEL.md. It permits use and redistribution, including commercial, but restricts third-party hosting (API/hosted inference) without prior written approval from the licensor, and gates very-large-scale product use. This repository is a weight download only — no inference endpoint is enabled.

Model by Bodhan AI (IITM Bodhan-AI Foundation), Centre of Excellence for AI in Education, Ministry of Education, Govt. of India. Quantization by Bonenk.

Downloads last month: 184

GGUF

Model size

7B params

Architecture

gemma4

Hardware compatibility

4-bit

6-bit

8-bit

16-bit

Model tree for Bonenk/indic-translate-GGUF

Base model

google/gemma-4-E4B

Finetuned

google/gemma-4-E4B-it

Finetuned

bodhan-ai/indic-translate

Quantized

(1)

this model