Instructions to use RoundRobin237/Gemma4-E2B-Dflash2-Android with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use RoundRobin237/Gemma4-E2B-Dflash2-Android 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 RoundRobin237/Gemma4-E2B-Dflash2-Android:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf RoundRobin237/Gemma4-E2B-Dflash2-Android:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf RoundRobin237/Gemma4-E2B-Dflash2-Android:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf RoundRobin237/Gemma4-E2B-Dflash2-Android: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 RoundRobin237/Gemma4-E2B-Dflash2-Android:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf RoundRobin237/Gemma4-E2B-Dflash2-Android: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 RoundRobin237/Gemma4-E2B-Dflash2-Android:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf RoundRobin237/Gemma4-E2B-Dflash2-Android:Q4_K_M

Use Docker

docker model run hf.co/RoundRobin237/Gemma4-E2B-Dflash2-Android:Q4_K_M

Llama new
LM Studio
Jan

vLLM

How to use RoundRobin237/Gemma4-E2B-Dflash2-Android with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "RoundRobin237/Gemma4-E2B-Dflash2-Android"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "RoundRobin237/Gemma4-E2B-Dflash2-Android",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Use Docker

docker model run hf.co/RoundRobin237/Gemma4-E2B-Dflash2-Android:Q4_K_M

Ollama
How to use RoundRobin237/Gemma4-E2B-Dflash2-Android with Ollama:
```
ollama run hf.co/RoundRobin237/Gemma4-E2B-Dflash2-Android:Q4_K_M
```
Unsloth Desktop

How to use RoundRobin237/Gemma4-E2B-Dflash2-Android with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf RoundRobin237/Gemma4-E2B-Dflash2-Android: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": "RoundRobin237/Gemma4-E2B-Dflash2-Android:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use RoundRobin237/Gemma4-E2B-Dflash2-Android with Docker Model Runner:
```
docker model run hf.co/RoundRobin237/Gemma4-E2B-Dflash2-Android:Q4_K_M
```

Lemonade

How to use RoundRobin237/Gemma4-E2B-Dflash2-Android with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull RoundRobin237/Gemma4-E2B-Dflash2-Android:Q4_K_M

Run and chat with the model

lemonade run user.Gemma4-E2B-Dflash2-Android-Q4_K_M

List all available models

lemonade list

Hermes Agent

How to use RoundRobin237/Gemma4-E2B-Dflash2-Android with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf RoundRobin237/Gemma4-E2B-Dflash2-Android: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 RoundRobin237/Gemma4-E2B-Dflash2-Android:Q4_K_M

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use RoundRobin237/Gemma4-E2B-Dflash2-Android with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf RoundRobin237/Gemma4-E2B-Dflash2-Android: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 "RoundRobin237/Gemma4-E2B-Dflash2-Android: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"

Gemma4-E2B-Dflash2-Android

File size: 2,028 Bytes

77e134b

{
  "base_model": "google/gemma-4-E2B-it",
  "base_revision": "3e22461f65e89153144f8adb70e3b8c2cc9845a7",
  "engine_git": "https://github.com/pytorch/executorch",
  "engine_revision": "ca56e1ffdba1c902d9202021d7297982d3501439",
  "qnn_sdk": "2.37.0.250724",
  "target_soc": "SM8450",
  "hexagon_arch": "v69",
  "context": 2048,
  "quantization": "mixed W4 block32 FFN gate/up/down; all other decoder projections W8 per-channel; final 8 cross-decoder FFN down projections restored to W8; A16; KV16; A16 x W8 attention boundary; Q4 group32 portable embeddings",
  "weight_bits": 4,
  "weight_granularity": "block",
  "block_size": 32,
  "w4_ffn_only": true,
  "w4_ffn_down": true,
  "w8_ffn_down_tail": 8,
  "compiled_shards": 11,
  "qnn_delegate_count": 12,
  "largest_delegate_bytes": 504553728,
  "source_calibration_context": 2048,
  "position_table_audit": true,
  "model_name": "Gemma4-E2B-Dflash2-Android",
  "original_baseline_pte_sha256": "38faf5d2337d9fb2b75b2ddca0fb3c63d916b2795c55bff6a1b6b92c027e8c7b",
  "original_baseline_pte_bytes": 3361616512,
  "original_baseline_pte_name": "gemma4-e2b-v69-mixed4b32-ffn-all-tail8w8-kv16-quality-2k-shard11.pte",
  "dflash2_target": {
    "name": "gemma4-e2b-dflash2-verify-2k.pte",
    "bytes": 3361310336,
    "sha256": "c05fe24cb87fe20f3c2fdf3bb942755486f4c34785ad4d28a5b5546a7378983b",
    "features": [
      3,
      13,
      23,
      33
    ],
    "verification_width": 4,
    "target_weights_changed_for_feature_export": false,
    "context_loading_order": "Head context loaded first; execution order unchanged"
  },
  "vision_target": {
    "name": "gemma4-e2b-dflash2-vision-verify-2k.pte",
    "sha256": "6475a64ba4c82a85dd635bde40cc51ece8effe47366565a1c6e58d48862cdbba",
    "source_sha256": "c05fe24cb87fe20f3c2fdf3bb942755486f4c34785ad4d28a5b5546a7378983b",
    "weights_and_QNN_contexts_unchanged": true,
    "modification": "Portable embedding operator descriptor alias only",
    "device_validated": true
  }
}