Instructions to use startlux-models/StartLux-Decision-v1-2B-BF16-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 startlux-models/StartLux-Decision-v1-2B-BF16-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 startlux-models/StartLux-Decision-v1-2B-BF16-GGUF:BF16
# Run inference directly in the terminal:
llama cli -hf startlux-models/StartLux-Decision-v1-2B-BF16-GGUF:BF16

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf startlux-models/StartLux-Decision-v1-2B-BF16-GGUF:BF16
# Run inference directly in the terminal:
llama cli -hf startlux-models/StartLux-Decision-v1-2B-BF16-GGUF:BF16

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 startlux-models/StartLux-Decision-v1-2B-BF16-GGUF:BF16
# Run inference directly in the terminal:
./llama-cli -hf startlux-models/StartLux-Decision-v1-2B-BF16-GGUF:BF16

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 startlux-models/StartLux-Decision-v1-2B-BF16-GGUF:BF16
# Run inference directly in the terminal:
./build/bin/llama-cli -hf startlux-models/StartLux-Decision-v1-2B-BF16-GGUF:BF16

Use Docker

docker model run hf.co/startlux-models/StartLux-Decision-v1-2B-BF16-GGUF:BF16

LM Studio
Jan
Ollama
How to use startlux-models/StartLux-Decision-v1-2B-BF16-GGUF with Ollama:
```
ollama run hf.co/startlux-models/StartLux-Decision-v1-2B-BF16-GGUF:BF16
```
Unsloth Desktop

How to use startlux-models/StartLux-Decision-v1-2B-BF16-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf startlux-models/StartLux-Decision-v1-2B-BF16-GGUF:BF16

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": "startlux-models/StartLux-Decision-v1-2B-BF16-GGUF:BF16"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use startlux-models/StartLux-Decision-v1-2B-BF16-GGUF with Docker Model Runner:
```
docker model run hf.co/startlux-models/StartLux-Decision-v1-2B-BF16-GGUF:BF16
```

Lemonade

How to use startlux-models/StartLux-Decision-v1-2B-BF16-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull startlux-models/StartLux-Decision-v1-2B-BF16-GGUF:BF16

Run and chat with the model

lemonade run user.StartLux-Decision-v1-2B-BF16-GGUF-BF16

List all available models

lemonade list

Hermes Agent

How to use startlux-models/StartLux-Decision-v1-2B-BF16-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 startlux-models/StartLux-Decision-v1-2B-BF16-GGUF:BF16

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 startlux-models/StartLux-Decision-v1-2B-BF16-GGUF:BF16

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use startlux-models/StartLux-Decision-v1-2B-BF16-GGUF with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf startlux-models/StartLux-Decision-v1-2B-BF16-GGUF:BF16

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 "startlux-models/StartLux-Decision-v1-2B-BF16-GGUF:BF16" \
  --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"

Lie24 commited on 5 days ago

Commit

183d616

verified ·

1 Parent(s): dec6ea4

Rename to StartLux-Decision-v1

Browse files

Files changed (1) hide show

README.md +10 -10

README.md CHANGED Viewed

@@ -1,6 +1,6 @@
 ---
 license: cc-by-nc-4.0
-base_model: startlux-models/StartLux-Decision-2B
 base_model_relation: quantized
 tags:
 - gguf
@@ -9,9 +9,9 @@ tags:
 - typed-decisions
 ---
-# StartLux-Decision-2B-BF16-GGUF
-[StartLux-Decision-2B](https://huggingface.co/startlux-models/StartLux-Decision-2B) in BF16 (16-bit) as a GGUF file for
 llama.cpp. The decision server in this repository turns it into the same typed decisions, with a probability for every
 option, as the original model.
@@ -21,10 +21,10 @@ option, as the original model.
 | Repository | Bits | Size | Same decision as the original | JevBench public, of 231 | |
 |:---:|:---:|:---:|:---:|:---:|:---:|
-| [StartLux-Decision-2B-Q8_0-GGUF](https://huggingface.co/startlux-models/StartLux-Decision-2B-Q8_0-GGUF) | 8-bit | 2.01 GB | 99.1% | 193 (83.5%) | recommended |
-| [StartLux-Decision-2B-Q4_K_M-GGUF](https://huggingface.co/startlux-models/StartLux-Decision-2B-Q4_K_M-GGUF) | 4-bit | 1.27 GB | 94.4% | 199 (86.1%) | smallest; at this size it changes more answers, prefer Q8_0 |
-| **StartLux-Decision-2B-BF16-GGUF** (this repository) | 16-bit | 3.78 GB | 100.0% | 195 (84.4%) | the original weights, unchanged |
-| [StartLux-Decision-2B](https://huggingface.co/startlux-models/StartLux-Decision-2B) (original weights) | | | | 195 (84.4%) | for comparison |
 </div>
@@ -35,8 +35,8 @@ are llama.cpp's standard quantizations of BF16, without an importance matrix. At
 ## Download and run
 ```bash
-hf download startlux-models/StartLux-Decision-2B-BF16-GGUF --local-dir StartLux-Decision-2B-BF16-GGUF
-cd StartLux-Decision-2B-BF16-GGUF
 pip install -r requirements.txt                     # transformers and torch; a CPU build of torch is enough
 # llama.cpp serves the weights; the decision server puts the prompt format, readout and calibration on top
@@ -70,4 +70,4 @@ in `gguf_server.py`, not in the weights. The file holds the text decoder only.
 The model weights are released under [CC BY-NC 4.0](https://creativecommons.org/licenses/by-nc/4.0/): free for
 research and other non-commercial use, with attribution. Commercial use requires a separate license from
 StartLux Labs; contact [contact@startlux.com](mailto:contact@startlux.com). The inference code in
-`startlux_decision/` is Apache-2.0. See [LICENSE](https://huggingface.co/startlux-models/StartLux-Decision-2B-BF16-GGUF/blob/main/LICENSE) and [NOTICE](https://huggingface.co/startlux-models/StartLux-Decision-2B-BF16-GGUF/blob/main/NOTICE).

 ---
 license: cc-by-nc-4.0
+base_model: startlux-models/StartLux-Decision-v1-2B
 base_model_relation: quantized
 tags:
 - gguf
 - typed-decisions
 ---
+# StartLux-Decision-v1-2B-BF16-GGUF
+[StartLux-Decision-v1-2B](https://huggingface.co/startlux-models/StartLux-Decision-v1-2B) in BF16 (16-bit) as a GGUF file for
 llama.cpp. The decision server in this repository turns it into the same typed decisions, with a probability for every
 option, as the original model.
 | Repository | Bits | Size | Same decision as the original | JevBench public, of 231 | |
 |:---:|:---:|:---:|:---:|:---:|:---:|
+| [StartLux-Decision-v1-2B-Q8_0-GGUF](https://huggingface.co/startlux-models/StartLux-Decision-v1-2B-Q8_0-GGUF) | 8-bit | 2.01 GB | 99.1% | 193 (83.5%) | recommended |
+| [StartLux-Decision-v1-2B-Q4_K_M-GGUF](https://huggingface.co/startlux-models/StartLux-Decision-v1-2B-Q4_K_M-GGUF) | 4-bit | 1.27 GB | 94.4% | 199 (86.1%) | smallest; at this size it changes more answers, prefer Q8_0 |
+| **StartLux-Decision-v1-2B-BF16-GGUF** (this repository) | 16-bit | 3.78 GB | 100.0% | 195 (84.4%) | the original weights, unchanged |
+| [StartLux-Decision-v1-2B](https://huggingface.co/startlux-models/StartLux-Decision-v1-2B) (original weights) | | | | 195 (84.4%) | for comparison |
 </div>
 ## Download and run
 ```bash
+hf download startlux-models/StartLux-Decision-v1-2B-BF16-GGUF --local-dir StartLux-Decision-v1-2B-BF16-GGUF
+cd StartLux-Decision-v1-2B-BF16-GGUF
 pip install -r requirements.txt                     # transformers and torch; a CPU build of torch is enough
 # llama.cpp serves the weights; the decision server puts the prompt format, readout and calibration on top
 The model weights are released under [CC BY-NC 4.0](https://creativecommons.org/licenses/by-nc/4.0/): free for
 research and other non-commercial use, with attribution. Commercial use requires a separate license from
 StartLux Labs; contact [contact@startlux.com](mailto:contact@startlux.com). The inference code in
+`startlux_decision/` is Apache-2.0. See [LICENSE](https://huggingface.co/startlux-models/StartLux-Decision-v1-2B-BF16-GGUF/blob/main/LICENSE) and [NOTICE](https://huggingface.co/startlux-models/StartLux-Decision-v1-2B-BF16-GGUF/blob/main/NOTICE).