Instructions to use OpenTARS/Gargantua-Code-Gen2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use OpenTARS/Gargantua-Code-Gen2 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 OpenTARS/Gargantua-Code-Gen2:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf OpenTARS/Gargantua-Code-Gen2:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf OpenTARS/Gargantua-Code-Gen2:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf OpenTARS/Gargantua-Code-Gen2:Q4_K_M

Build from source code

git clone https://github.com/ggml-org/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-app
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama serve -hf OpenTARS/Gargantua-Code-Gen2:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama cli -hf OpenTARS/Gargantua-Code-Gen2:Q4_K_M

Use Docker

docker model run hf.co/OpenTARS/Gargantua-Code-Gen2:Q4_K_M

Llama new
LM Studio
Jan
Ollama
How to use OpenTARS/Gargantua-Code-Gen2 with Ollama:
```
ollama run hf.co/OpenTARS/Gargantua-Code-Gen2:Q4_K_M
```
Unsloth Desktop

How to use OpenTARS/Gargantua-Code-Gen2 with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf OpenTARS/Gargantua-Code-Gen2: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": "OpenTARS/Gargantua-Code-Gen2:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use OpenTARS/Gargantua-Code-Gen2 with Docker Model Runner:
```
docker model run hf.co/OpenTARS/Gargantua-Code-Gen2:Q4_K_M
```

Lemonade

How to use OpenTARS/Gargantua-Code-Gen2 with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull OpenTARS/Gargantua-Code-Gen2:Q4_K_M

Run and chat with the model

lemonade run user.Gargantua-Code-Gen2-Q4_K_M

List all available models

lemonade list

Hermes Agent

How to use OpenTARS/Gargantua-Code-Gen2 with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf OpenTARS/Gargantua-Code-Gen2: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 OpenTARS/Gargantua-Code-Gen2:Q4_K_M

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use OpenTARS/Gargantua-Code-Gen2 with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf OpenTARS/Gargantua-Code-Gen2: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 "OpenTARS/Gargantua-Code-Gen2: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"

ProjectOpenTARS4 commited on 2 days ago

Commit

1675b49

verified ·

1 Parent(s): fe92ea1

Reword limits

Browse files

Files changed (1) hide show

README.md +1 -1

README.md CHANGED Viewed

@@ -30,7 +30,7 @@ No dangerous actions. Average 7.7 s per task on an RTX 3050.
 ## Limits
-- **Reading code is weak (1/10):** in 6 of the 9 misses the model returned nothing. We do not know the cause yet.
 - The tasks are simple (create a file, fix a short script). This is not a general programming benchmark.
 - The untrained base (`qwen2.5-coder:7b`) scores 0/60 on the same test, because it does not know the openTARS prompt and tools. That shows what the training teaches, not that the base is bad at code.
 - The Q4_K_M export has not been compared against the 16-bit model.

 ## Limits
+- Reading a script and saying what it prints scored 1 of 10: in 6 of the 9 misses the model returned nothing. We do not know the cause yet.
 - The tasks are simple (create a file, fix a short script). This is not a general programming benchmark.
 - The untrained base (`qwen2.5-coder:7b`) scores 0/60 on the same test, because it does not know the openTARS prompt and tools. That shows what the training teaches, not that the base is bad at code.
 - The Q4_K_M export has not been compared against the 16-bit model.