Instructions to use SciTools/OnBoard with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use SciTools/OnBoard 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 SciTools/OnBoard:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf SciTools/OnBoard:Q4_K_M

Install from WinGet (Windows)

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

Use Docker

docker model run hf.co/SciTools/OnBoard:Q4_K_M

LM Studio
Jan
Ollama
How to use SciTools/OnBoard with Ollama:
```
ollama run hf.co/SciTools/OnBoard:Q4_K_M
```
Unsloth Desktop

How to use SciTools/OnBoard with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf SciTools/OnBoard: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": "SciTools/OnBoard:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use SciTools/OnBoard with Docker Model Runner:
```
docker model run hf.co/SciTools/OnBoard:Q4_K_M
```

Lemonade

How to use SciTools/OnBoard with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull SciTools/OnBoard:Q4_K_M

Run and chat with the model

lemonade run user.OnBoard-Q4_K_M

List all available models

lemonade list

Hermes Agent

How to use SciTools/OnBoard with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf SciTools/OnBoard: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 SciTools/OnBoard:Q4_K_M

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use SciTools/OnBoard with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf SciTools/OnBoard: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 "SciTools/OnBoard: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"

zenpeach commited on 18 days ago

Commit

de12707

verified ·

1 Parent(s): c772b74

README: Gemma 4 E4B now ships the QAT UD-Q4_K_XL build

Browse files

Files changed (1) hide show

README.md +7 -3

README.md CHANGED Viewed

@@ -35,7 +35,7 @@ the best chat answers.
 | Qwen3.6-35B-A3B | 22.4 GB | Highly accurate code summaries, and much faster than other models this large. |
 | Gemma 4 12B | 7.1 GB | The best chat answers of any model we tested. |
 | Qwen3.5 9B | 5.7 GB | Did well in chat testing, and digs a little deeper into code before answering. |
-| Gemma 4 E4B | 5.0 GB | Did well in our chat testing. |
 | Gemma 4 E2B | 3.1 GB | OnBoard's default model. Good at both chat and code summaries. |
 | Qwen3.5 2B | 1.3 GB | Writes good code summaries remarkably fast, but struggled in our chat testing. |
@@ -49,12 +49,16 @@ the best chat answers.
 | Qwen3.6-35B-A3B-UD-Q4_K_XL.gguf | Qwen3.6-35B-A3B | Alibaba (China) | Apache-2.0 | [unsloth/Qwen3.6-35B-A3B-GGUF](https://huggingface.co/unsloth/Qwen3.6-35B-A3B-GGUF) |
 | gemma-4-12b-it-Q4_K_M.gguf | Gemma 4 12B | Google (United States) | Apache-2.0 | [unsloth/gemma-4-12b-it-GGUF](https://huggingface.co/unsloth/gemma-4-12b-it-GGUF) |
 | Qwen3.5-9B-Q4_K_M.gguf | Qwen3.5 9B | Alibaba (China) | Apache-2.0 | [unsloth/Qwen3.5-9B-GGUF](https://huggingface.co/unsloth/Qwen3.5-9B-GGUF) |
-| gemma-4-E4B-it-Q4_K_M.gguf | Gemma 4 E4B | Google (United States) | Apache-2.0 | [unsloth/gemma-4-E4B-it-GGUF](https://huggingface.co/unsloth/gemma-4-E4B-it-GGUF) |
 | gemma-4-E2B-it-Q4_K_M.gguf | Gemma 4 E2B | Google (United States) | Apache-2.0 | [unsloth/gemma-4-E2B-it-GGUF](https://huggingface.co/unsloth/gemma-4-E2B-it-GGUF) |
 | Qwen3.5-2B-Q4_K_M.gguf | Qwen3.5 2B | Alibaba (China) | Apache-2.0 | [unsloth/Qwen3.5-2B-GGUF](https://huggingface.co/unsloth/Qwen3.5-2B-GGUF) |
 Quantization policy: Q4_K_M for models ≤12B; Unsloth Dynamic (UD-Q4_K_XL) for
-the large MoE, where the dynamic quant's quality edge matters most. gpt-oss-120b
 is a two-part split: download both parts into the same directory and point the
 server at part 1 — llama.cpp finds the second part automatically.

 | Qwen3.6-35B-A3B | 22.4 GB | Highly accurate code summaries, and much faster than other models this large. |
 | Gemma 4 12B | 7.1 GB | The best chat answers of any model we tested. |
 | Qwen3.5 9B | 5.7 GB | Did well in chat testing, and digs a little deeper into code before answering. |
+| Gemma 4 E4B | 4.2 GB | Did well in our chat testing. |
 | Gemma 4 E2B | 3.1 GB | OnBoard's default model. Good at both chat and code summaries. |
 | Qwen3.5 2B | 1.3 GB | Writes good code summaries remarkably fast, but struggled in our chat testing. |
 | Qwen3.6-35B-A3B-UD-Q4_K_XL.gguf | Qwen3.6-35B-A3B | Alibaba (China) | Apache-2.0 | [unsloth/Qwen3.6-35B-A3B-GGUF](https://huggingface.co/unsloth/Qwen3.6-35B-A3B-GGUF) |
 | gemma-4-12b-it-Q4_K_M.gguf | Gemma 4 12B | Google (United States) | Apache-2.0 | [unsloth/gemma-4-12b-it-GGUF](https://huggingface.co/unsloth/gemma-4-12b-it-GGUF) |
 | Qwen3.5-9B-Q4_K_M.gguf | Qwen3.5 9B | Alibaba (China) | Apache-2.0 | [unsloth/Qwen3.5-9B-GGUF](https://huggingface.co/unsloth/Qwen3.5-9B-GGUF) |
+| gemma-4-E4B-it-qat-UD-Q4_K_XL.gguf | Gemma 4 E4B | Google (United States) | Apache-2.0 | [unsloth/gemma-4-E4B-it-qat-GGUF](https://huggingface.co/unsloth/gemma-4-E4B-it-qat-GGUF) |
 | gemma-4-E2B-it-Q4_K_M.gguf | Gemma 4 E2B | Google (United States) | Apache-2.0 | [unsloth/gemma-4-E2B-it-GGUF](https://huggingface.co/unsloth/gemma-4-E2B-it-GGUF) |
 | Qwen3.5-2B-Q4_K_M.gguf | Qwen3.5 2B | Alibaba (China) | Apache-2.0 | [unsloth/Qwen3.5-2B-GGUF](https://huggingface.co/unsloth/Qwen3.5-2B-GGUF) |
 Quantization policy: Q4_K_M for models ≤12B; Unsloth Dynamic (UD-Q4_K_XL) for
+the large MoE, where the dynamic quant's quality edge matters most. Gemma 4 E4B
+is the exception: it uses Google's quantization-aware-trained (QAT) build in
+UD-Q4_K_XL, which qualified on OnBoard's chat test in 3 of 3 runs and is smaller
+than the Q4_K_M. The earlier `gemma-4-E4B-it-Q4_K_M.gguf` stays here for
+installs that already use it. gpt-oss-120b
 is a two-part split: download both parts into the same directory and point the
 server at part 1 — llama.cpp finds the second part automatically.