Instructions to use prithivMLmods/SFT-4B-ScaleCUA-MixedOpen-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use prithivMLmods/SFT-4B-ScaleCUA-MixedOpen-GGUF with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("image-text-to-text", model="prithivMLmods/SFT-4B-ScaleCUA-MixedOpen-GGUF")

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("prithivMLmods/SFT-4B-ScaleCUA-MixedOpen-GGUF", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use prithivMLmods/SFT-4B-ScaleCUA-MixedOpen-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 prithivMLmods/SFT-4B-ScaleCUA-MixedOpen-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf prithivMLmods/SFT-4B-ScaleCUA-MixedOpen-GGUF:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf prithivMLmods/SFT-4B-ScaleCUA-MixedOpen-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf prithivMLmods/SFT-4B-ScaleCUA-MixedOpen-GGUF: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 prithivMLmods/SFT-4B-ScaleCUA-MixedOpen-GGUF:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf prithivMLmods/SFT-4B-ScaleCUA-MixedOpen-GGUF: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 prithivMLmods/SFT-4B-ScaleCUA-MixedOpen-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf prithivMLmods/SFT-4B-ScaleCUA-MixedOpen-GGUF:Q4_K_M

Use Docker

docker model run hf.co/prithivMLmods/SFT-4B-ScaleCUA-MixedOpen-GGUF:Q4_K_M

LM Studio
Jan

vLLM

How to use prithivMLmods/SFT-4B-ScaleCUA-MixedOpen-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "prithivMLmods/SFT-4B-ScaleCUA-MixedOpen-GGUF"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "prithivMLmods/SFT-4B-ScaleCUA-MixedOpen-GGUF",
		"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/prithivMLmods/SFT-4B-ScaleCUA-MixedOpen-GGUF:Q4_K_M

SGLang

How to use prithivMLmods/SFT-4B-ScaleCUA-MixedOpen-GGUF with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "prithivMLmods/SFT-4B-ScaleCUA-MixedOpen-GGUF" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "prithivMLmods/SFT-4B-ScaleCUA-MixedOpen-GGUF",
		"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 images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "prithivMLmods/SFT-4B-ScaleCUA-MixedOpen-GGUF" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "prithivMLmods/SFT-4B-ScaleCUA-MixedOpen-GGUF",
		"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"
						}
					}
				]
			}
		]
	}'

Ollama
How to use prithivMLmods/SFT-4B-ScaleCUA-MixedOpen-GGUF with Ollama:
```
ollama run hf.co/prithivMLmods/SFT-4B-ScaleCUA-MixedOpen-GGUF:Q4_K_M
```
Unsloth Desktop

How to use prithivMLmods/SFT-4B-ScaleCUA-MixedOpen-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf prithivMLmods/SFT-4B-ScaleCUA-MixedOpen-GGUF: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": "prithivMLmods/SFT-4B-ScaleCUA-MixedOpen-GGUF:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use prithivMLmods/SFT-4B-ScaleCUA-MixedOpen-GGUF with Docker Model Runner:
```
docker model run hf.co/prithivMLmods/SFT-4B-ScaleCUA-MixedOpen-GGUF:Q4_K_M
```

Lemonade

How to use prithivMLmods/SFT-4B-ScaleCUA-MixedOpen-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull prithivMLmods/SFT-4B-ScaleCUA-MixedOpen-GGUF:Q4_K_M

Run and chat with the model

lemonade run user.SFT-4B-ScaleCUA-MixedOpen-GGUF-Q4_K_M

List all available models

lemonade list

Hermes Agent

How to use prithivMLmods/SFT-4B-ScaleCUA-MixedOpen-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 prithivMLmods/SFT-4B-ScaleCUA-MixedOpen-GGUF: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 prithivMLmods/SFT-4B-ScaleCUA-MixedOpen-GGUF:Q4_K_M

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use prithivMLmods/SFT-4B-ScaleCUA-MixedOpen-GGUF with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf prithivMLmods/SFT-4B-ScaleCUA-MixedOpen-GGUF: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 "prithivMLmods/SFT-4B-ScaleCUA-MixedOpen-GGUF: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"

prithivMLmods commited on 25 days ago

Commit

de18af6

verified ·

1 Parent(s): 967ef16

Update README.md

Browse files

Files changed (1) hide show

README.md +1 -1

README.md CHANGED Viewed

@@ -17,7 +17,7 @@ pipeline_tag: image-text-to-text
 # **SFT-4B-ScaleCUA-MixedOpen-GGUF**
-> **SFT-4B-ScaleCUA-MixedOpen** is a Qwen3-VL-4B-Instruct checkpoint supervised-finetuned on 1,430 Lite.ScaleCUA trajectories pooled in equal proportion from three open-source teacher agents — Qwen3.8-27B (478 trajectories), Qwen3.5-27B (477), and EvoCUA-8B-20260105 (475), one trajectory per task, 1,390 of which are full successes — trained via cua-lite + slime with token-level SFT on assistant-action tokens over 3 epochs (1,070 steps) on 2x 80GB H100 GPUs. This epoch-3 checkpoint (`iter_1070`) is the strongest of the measured ScaleCUA SFT arms on the Lite.OSWorld `eval` split (332 tasks, greedy decoding, one host/protocol), reaching a mean episode return of 0.3927 (125/332 success) — beating the best single-teacher arm (Qwen3.5-27B, 0.3682) by +0.0245/+8 tasks and the closest-matched single-teacher arm (Qwen38, 0.3623) by +0.0304/+10 tasks, both clearing the paper's >0.02 mean and ≥7 task threshold for a real effect, though the authors caution this isn't a clean single-variable ablation since teacher identity, trajectory count, and task coverage all move together. Notably, doubling trajectories per task (`-MixedOpen-cap2`, 2,520 trajectories) actually underperforms this model by −0.0125/−5 tasks, suggesting broader task coverage matters more than redundant per-task examples, and per-domain results show `multi_apps` coordination (28% of tasks) remains the weakest capability across all teacher mixes at just 0.159 mean return. The model requires the `qwen3_vl` adapter config with `full_history_size=4` for correct serving, since it was trained on that specific history-rendering protocol.
 ## Model Files

 # **SFT-4B-ScaleCUA-MixedOpen-GGUF**
+> **[SFT-4B-ScaleCUA-MixedOpen](https://huggingface.co/HaoranLiu/SFT-4B-ScaleCUA-MixedOpen)** is a Qwen3-VL-4B-Instruct checkpoint supervised-finetuned on 1,430 Lite.ScaleCUA trajectories pooled in equal proportion from three open-source teacher agents — Qwen3.8-27B (478 trajectories), Qwen3.5-27B (477), and EvoCUA-8B-20260105 (475), one trajectory per task, 1,390 of which are full successes — trained via cua-lite + slime with token-level SFT on assistant-action tokens over 3 epochs (1,070 steps) on 2x 80GB H100 GPUs. This epoch-3 checkpoint (`iter_1070`) is the strongest of the measured ScaleCUA SFT arms on the Lite.OSWorld `eval` split (332 tasks, greedy decoding, one host/protocol), reaching a mean episode return of 0.3927 (125/332 success) — beating the best single-teacher arm (Qwen3.5-27B, 0.3682) by +0.0245/+8 tasks and the closest-matched single-teacher arm (Qwen38, 0.3623) by +0.0304/+10 tasks, both clearing the paper's >0.02 mean and ≥7 task threshold for a real effect, though the authors caution this isn't a clean single-variable ablation since teacher identity, trajectory count, and task coverage all move together. Notably, doubling trajectories per task (`-MixedOpen-cap2`, 2,520 trajectories) actually underperforms this model by −0.0125/−5 tasks, suggesting broader task coverage matters more than redundant per-task examples, and per-domain results show `multi_apps` coordination (28% of tasks) remains the weakest capability across all teacher mixes at just 0.159 mean return. The model requires the `qwen3_vl` adapter config with `full_history_size=4` for correct serving, since it was trained on that specific history-rendering protocol.
 ## Model Files