Instructions to use ef-dai-team/gemma-4-E4B-wallet-ft with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries
PEFT
How to use ef-dai-team/gemma-4-E4B-wallet-ft with PEFT:
```
Task type is invalid.
```
Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use ef-dai-team/gemma-4-E4B-wallet-ft 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 ef-dai-team/gemma-4-E4B-wallet-ft:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf ef-dai-team/gemma-4-E4B-wallet-ft:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf ef-dai-team/gemma-4-E4B-wallet-ft:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf ef-dai-team/gemma-4-E4B-wallet-ft: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 ef-dai-team/gemma-4-E4B-wallet-ft:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf ef-dai-team/gemma-4-E4B-wallet-ft: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 ef-dai-team/gemma-4-E4B-wallet-ft:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf ef-dai-team/gemma-4-E4B-wallet-ft:Q4_K_M

Use Docker

docker model run hf.co/ef-dai-team/gemma-4-E4B-wallet-ft:Q4_K_M

LM Studio
Jan

vLLM

How to use ef-dai-team/gemma-4-E4B-wallet-ft with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "ef-dai-team/gemma-4-E4B-wallet-ft"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ef-dai-team/gemma-4-E4B-wallet-ft",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/ef-dai-team/gemma-4-E4B-wallet-ft:Q4_K_M

Ollama
How to use ef-dai-team/gemma-4-E4B-wallet-ft with Ollama:
```
ollama run hf.co/ef-dai-team/gemma-4-E4B-wallet-ft:Q4_K_M
```

Unsloth Studio

How to use ef-dai-team/gemma-4-E4B-wallet-ft 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 ef-dai-team/gemma-4-E4B-wallet-ft 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 ef-dai-team/gemma-4-E4B-wallet-ft to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for ef-dai-team/gemma-4-E4B-wallet-ft to start chatting

How to use ef-dai-team/gemma-4-E4B-wallet-ft with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf ef-dai-team/gemma-4-E4B-wallet-ft: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": "ef-dai-team/gemma-4-E4B-wallet-ft:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

OpenClaw new

How to use ef-dai-team/gemma-4-E4B-wallet-ft with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf ef-dai-team/gemma-4-E4B-wallet-ft: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 "ef-dai-team/gemma-4-E4B-wallet-ft: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 ef-dai-team/gemma-4-E4B-wallet-ft with Docker Model Runner:
```
docker model run hf.co/ef-dai-team/gemma-4-E4B-wallet-ft:Q4_K_M
```

Lemonade

How to use ef-dai-team/gemma-4-E4B-wallet-ft with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull ef-dai-team/gemma-4-E4B-wallet-ft:Q4_K_M

Run and chat with the model

lemonade run user.gemma-4-E4B-wallet-ft-Q4_K_M

List all available models

lemonade list

Hermes Agent

How to use ef-dai-team/gemma-4-E4B-wallet-ft with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf ef-dai-team/gemma-4-E4B-wallet-ft: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 ef-dai-team/gemma-4-E4B-wallet-ft:Q4_K_M

Run Hermes

hermes

Atomic Chat

gemma-4-E4B-wallet-ft

A LoRA fine-tune of Gemma-4 E4B that turns a natural-language wallet request into the exact structured tool call a macOS Ethereum wallet can execute — resolving ENS names, selecting from five tools, and converting human amounts to base units.

"Send 0.1 ETH to vitalik.eth"

[{"name": "executeTx",
  "arguments": {"chainId": "1",
                "to": "0xd8dA6BF26964aF9D7eEd9e03E53415D37aA96045",
                "value": "100000000000000000",
                "function": null, "args": []}}]

Results

On the 307-case evaluation set, scored by exact match on every field of every call:

Model	Score
Gemma-4 E4B base (what the wallet ships today)	9.8%
This fine-tune	80.1%
gpt-5 (capable anchor)	96.7%

Category	Cases	Score
transfers	101	71.3%
swaps	96	83.3%
multi-turn	75	81.3%
ablation + refusal	35	94.3%

A 70-point lift over the shipped base model, on-device, at Q4_K_M.

gemma-4-E4B-wallet-ft.Q4_K_M.gguf — merged + quantised, 5.3 GB, runs via llama-cpp-python
adapter/ — the LoRA adapter (r=16, α=16) and training state

Usage

from llama_cpp import Llama
llm = Llama.from_pretrained(repo_id="ef-dai-team/gemma-4-E4B-wallet-ft",
                            filename="*Q4_K_M.gguf", n_ctx=4096)

Serve it with the system role (not developer) and the Gemma-4 delimiter dialect. The evaluation harness's provider handles both.

Training data

ef-dai-team/wallet-tool-calling-ft — 1739 SFT examples, disjoint from the evaluation set by construction, which is what makes the numbers above meaningful rather than memorisation.

The same recipe on a smaller model

Applied to FunctionGemma-270M the identical data and recipe produced no lift at all (8.1% → 8.8%, and 0% on every category requiring a call) — see ef-dai-team/functiongemma-270m-wallet-ft. Model capacity, not data, was the constraint.

Evaluation

Full per-case results, including this model's actual output on all 307 cases, are in the eval report Space.

License

Apache-2.0, matching the base model google/gemma-4-E4B-it. The training data is Apache-2.0 as well.

Downloads last month: 36

GGUF

Model size

8B params

Architecture

gemma4

Hardware compatibility

4-bit

Model tree for ef-dai-team/gemma-4-E4B-wallet-ft

Base model

google/gemma-4-E4B

Finetuned

google/gemma-4-E4B-it

Adapter

(311)

this model

ef-dai-team
/

gemma-4-E4B-wallet-ft

gemma-4-E4B-wallet-ft

Results

Contents

Usage

Training data

The same recipe on a smaller model

Evaluation

License

Model tree for ef-dai-team/gemma-4-E4B-wallet-ft