Instructions to use meetara-lab/meetara-gemma4-e2b-it-gguf with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use meetara-lab/meetara-gemma4-e2b-it-gguf with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="meetara-lab/meetara-gemma4-e2b-it-gguf",
	filename="meetara-gemma4-e2b-it-Q4_K_M.gguf",
)

llm.create_chat_completion(
	messages = [
		{
			"role": "user",
			"content": "What is the capital of France?"
		}
	]
)

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use meetara-lab/meetara-gemma4-e2b-it-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 meetara-lab/meetara-gemma4-e2b-it-gguf:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf meetara-lab/meetara-gemma4-e2b-it-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 meetara-lab/meetara-gemma4-e2b-it-gguf:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf meetara-lab/meetara-gemma4-e2b-it-gguf:Q4_K_M

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 meetara-lab/meetara-gemma4-e2b-it-gguf:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf meetara-lab/meetara-gemma4-e2b-it-gguf:Q4_K_M

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 meetara-lab/meetara-gemma4-e2b-it-gguf:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf meetara-lab/meetara-gemma4-e2b-it-gguf:Q4_K_M

Use Docker

docker model run hf.co/meetara-lab/meetara-gemma4-e2b-it-gguf:Q4_K_M

LM Studio
Jan

vLLM

How to use meetara-lab/meetara-gemma4-e2b-it-gguf with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "meetara-lab/meetara-gemma4-e2b-it-gguf"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "meetara-lab/meetara-gemma4-e2b-it-gguf",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/meetara-lab/meetara-gemma4-e2b-it-gguf:Q4_K_M

Ollama
How to use meetara-lab/meetara-gemma4-e2b-it-gguf with Ollama:
```
ollama run hf.co/meetara-lab/meetara-gemma4-e2b-it-gguf:Q4_K_M
```

Unsloth Studio

How to use meetara-lab/meetara-gemma4-e2b-it-gguf with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for meetara-lab/meetara-gemma4-e2b-it-gguf to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for meetara-lab/meetara-gemma4-e2b-it-gguf to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for meetara-lab/meetara-gemma4-e2b-it-gguf to start chatting

How to use meetara-lab/meetara-gemma4-e2b-it-gguf with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf meetara-lab/meetara-gemma4-e2b-it-gguf:Q4_K_M

Configure the model in Pi

# Install Pi:
npm install -g @mariozechner/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "meetara-lab/meetara-gemma4-e2b-it-gguf:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use meetara-lab/meetara-gemma4-e2b-it-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 meetara-lab/meetara-gemma4-e2b-it-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 meetara-lab/meetara-gemma4-e2b-it-gguf:Q4_K_M

Run Hermes

hermes

Atomic Chat new

OpenClaw new

How to use meetara-lab/meetara-gemma4-e2b-it-gguf with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf meetara-lab/meetara-gemma4-e2b-it-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 "meetara-lab/meetara-gemma4-e2b-it-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"

Docker Model Runner
How to use meetara-lab/meetara-gemma4-e2b-it-gguf with Docker Model Runner:
```
docker model run hf.co/meetara-lab/meetara-gemma4-e2b-it-gguf:Q4_K_M
```

Lemonade

How to use meetara-lab/meetara-gemma4-e2b-it-gguf with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull meetara-lab/meetara-gemma4-e2b-it-gguf:Q4_K_M

Run and chat with the model

lemonade run user.meetara-gemma4-e2b-it-gguf-Q4_K_M

List all available models

lemonade list

rameshbasina commited on 17 days ago

Commit

24e175a

verified ·

1 Parent(s): 55a080c

Upload README.md with huggingface_hub

Browse files

Files changed (1) hide show

README.md +33 -37

README.md CHANGED Viewed

@@ -7,61 +7,57 @@ tags:
   - meeTARA
   - gguf
   - instruction-tuned
 base_model: google/gemma-4-E2B-it
 pipeline_tag: text-generation
 ---
-# meetara-gemma4-e2b-it (meeTARA Gemma 4 distribution)
-This is a **meeTARA-branded export** of the official open-weights checkpoint **`google/gemma-4-E2B-it`**
-(Gemma 4, instruction-tuned). Gemma-native turn formatting is preserved, and meeTARA default
-system behavior is injected only when the caller does not provide a `system` message, so
-multimodal formatting, tool calling, and thinking modes stay compatible with Transformers and
-current `llama.cpp` converters.
-**me²TARA** = Mental & Emotional Empathetic Technology for Adaptive Response & Assistance.
-## meeTARA welcome and response structure (optional)
-This folder includes **`meetara_default_system_en.txt`**: the same **welcome + structured-response**
-intent as the v3 "simple" ChatML block, in **plain text** for you to pass as a **system** (or
-developer) message in your app or API. The same default content is also mirrored into Gemma-native
-template metadata so behavior remains present when no explicit `system` message is provided.
-## Why not the v3 "enhanced" ChatML template?
-`convert_to_gguf_meetara_v3_enhanced.py` injects a large Qwen-style **ChatML** chat template. Gemma 4
-expects Google's **`<|turn|>`** template and special tokens; mixing formats would produce broken
-prompts. For Gemma, use this script for GGUF packaging; use **`meetara_default_system_en.txt`** plus
-your app, or a **Gemma-native** fine-tune, if you need meeTARA behavior inside the stack.
-## GGUF build
-Built with repo `llama.cpp`:
-```bash
-python llama.cpp/convert_hf_to_gguf.py <this_folder> --outfile meetara-gemma4-e2b-it-Q4_K_M.gguf
-# optional vision/audio projector:
-python llama.cpp/convert_hf_to_gguf.py <this_folder> --mmproj --outfile meetara-vl-gemma4-e2b-it.gguf
-```
-## Vision-Language features (VL)
-This release is intended to be used as a pair:
-- Text model: `meetara-gemma4-e2b-it-Q4_K_M.gguf`
-- Vision projector: `meetara-vl-gemma4-e2b-it.gguf`
-How to use:
-- Use the text GGUF for normal chat/completion.
-- For image + text workflows, load both the text GGUF and the matching `meetara-vl-*.gguf` projector.
-- Keep the text model and projector from the same variant (E2B with E2B VL) to avoid mismatch issues.
-## License
-Apache 2.0 (same as upstream Gemma 4).
-## Credits
-Weights and tokenizer: Google DeepMind (`google/gemma-4-E2B-it`). Packaging: meeTARA project.

   - meeTARA
   - gguf
   - instruction-tuned
+  - multimodal
+  - vision-language
 base_model: google/gemma-4-E2B-it
 pipeline_tag: text-generation
 ---
+# meeTARA Gemma 4 E2B IT (GGUF)
+This repository provides the meeTARA GGUF distribution of **google/gemma-4-E2B-it** for local inference.
+**meeTARA** is an offline-first, privacy-focused, emotionally intelligent assistant focused on:
+- empathetic, practical assistance
+- clear action-oriented responses
+- safer guidance for sensitive topics
+- privacy-friendly local usage
+## High-level capabilities
+- Lightweight and efficient assistant model for local deployment
+- Strong instruction following for everyday tasks
+- Adaptive communication style (clear, concise, practical)
+- Faster response profile with lower memory requirements than larger variants
+- Optional image + text workflows using the matching VL projector
+## Files in this repo
+- `meetara-gemma4-e2b-it-Q4_K_M.gguf` (text model)
+- `meetara-vl-gemma4-e2b-it.gguf` (vision-language projector)
+## Vision-language (VL)
+For multimodal usage, load both files together:
+- text GGUF: `meetara-gemma4-e2b-it-Q4_K_M.gguf`
+- projector GGUF: `meetara-vl-gemma4-e2b-it.gguf`
+Keep the pair matched by variant (E2B text with E2B VL projector).
+## Intended use
+- local assistant experiences
+- emotional support style conversations (non-clinical)
+- productivity and everyday guidance
+- multimodal assistant workflows when paired with VL projector
+## Safety note
+This model is not a substitute for professional medical, legal, financial, or emergency advice. For high-risk situations, escalate to qualified professionals.
+## Base model and license
+- Base model: `google/gemma-4-E2B-it`
+- License: Apache 2.0 (same as upstream)