Instructions to use OfficerChul/InfiGUI-G1-3B-Android-Control-5a with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use OfficerChul/InfiGUI-G1-3B-Android-Control-5a with Transformers:

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

pipe = pipeline("image-text-to-text", model="OfficerChul/InfiGUI-G1-3B-Android-Control-5a")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
pipe(text=messages)

# Load model directly
from transformers import AutoProcessor, AutoModelForMultimodalLM

processor = AutoProcessor.from_pretrained("OfficerChul/InfiGUI-G1-3B-Android-Control-5a")
model = AutoModelForMultimodalLM.from_pretrained("OfficerChul/InfiGUI-G1-3B-Android-Control-5a", device_map="auto")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
inputs = processor.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use OfficerChul/InfiGUI-G1-3B-Android-Control-5a with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "OfficerChul/InfiGUI-G1-3B-Android-Control-5a"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "OfficerChul/InfiGUI-G1-3B-Android-Control-5a",
		"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/OfficerChul/InfiGUI-G1-3B-Android-Control-5a

SGLang

How to use OfficerChul/InfiGUI-G1-3B-Android-Control-5a 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 "OfficerChul/InfiGUI-G1-3B-Android-Control-5a" \
    --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": "OfficerChul/InfiGUI-G1-3B-Android-Control-5a",
		"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 "OfficerChul/InfiGUI-G1-3B-Android-Control-5a" \
        --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": "OfficerChul/InfiGUI-G1-3B-Android-Control-5a",
		"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"
						}
					}
				]
			}
		]
	}'

Docker Model Runner
How to use OfficerChul/InfiGUI-G1-3B-Android-Control-5a with Docker Model Runner:
```
docker model run hf.co/OfficerChul/InfiGUI-G1-3B-Android-Control-5a
```

InfiGUI-G1-3B-Android-Control-5a / trainer_log.jsonl

OfficerChul

Upload folder using huggingface_hub

2a98ab1 verified 11 months ago

Raw

History Blame Contribute Delete

9.52 kB

	{"current_steps": 10, "total_steps": 450, "loss": 2.8571, "lr": 4.000000000000001e-06, "epoch": 0.11196641007697691, "percentage": 2.22, "elapsed_time": "0:10:32", "remaining_time": "7:43:35"}
	{"current_steps": 20, "total_steps": 450, "loss": 0.6148, "lr": 8.444444444444446e-06, "epoch": 0.22393282015395383, "percentage": 4.44, "elapsed_time": "0:21:00", "remaining_time": "7:31:39"}
	{"current_steps": 30, "total_steps": 450, "loss": 0.3928, "lr": 1.288888888888889e-05, "epoch": 0.3358992302309307, "percentage": 6.67, "elapsed_time": "0:31:32", "remaining_time": "7:21:39"}
	{"current_steps": 40, "total_steps": 450, "loss": 0.3458, "lr": 1.7333333333333336e-05, "epoch": 0.44786564030790765, "percentage": 8.89, "elapsed_time": "0:42:07", "remaining_time": "7:11:42"}
	{"current_steps": 50, "total_steps": 450, "loss": 0.3267, "lr": 1.9995186678809513e-05, "epoch": 0.5598320503848845, "percentage": 11.11, "elapsed_time": "0:52:40", "remaining_time": "7:01:24"}
	{"current_steps": 60, "total_steps": 450, "loss": 0.3158, "lr": 1.9941090015469614e-05, "epoch": 0.6717984604618614, "percentage": 13.33, "elapsed_time": "1:03:14", "remaining_time": "6:51:02"}
	{"current_steps": 70, "total_steps": 450, "loss": 0.3016, "lr": 1.9827206467064133e-05, "epoch": 0.7837648705388384, "percentage": 15.56, "elapsed_time": "1:13:47", "remaining_time": "6:40:35"}
	{"current_steps": 80, "total_steps": 450, "loss": 0.278, "lr": 1.9654220942653223e-05, "epoch": 0.8957312806158153, "percentage": 17.78, "elapsed_time": "1:24:23", "remaining_time": "6:30:18"}
	{"current_steps": 90, "total_steps": 450, "loss": 0.2639, "lr": 1.9423173797534924e-05, "epoch": 1.0, "percentage": 20.0, "elapsed_time": "1:34:15", "remaining_time": "6:17:02"}
	{"current_steps": 100, "total_steps": 450, "loss": 0.2532, "lr": 1.913545457642601e-05, "epoch": 1.1119664100769768, "percentage": 22.22, "elapsed_time": "1:44:53", "remaining_time": "6:07:07"}
	{"current_steps": 100, "total_steps": 450, "eval_loss": 0.2309780865907669, "epoch": 1.1119664100769768, "percentage": 22.22, "elapsed_time": "1:50:28", "remaining_time": "6:26:40"}
	{"current_steps": 110, "total_steps": 450, "loss": 0.2366, "lr": 1.8792793656576544e-05, "epoch": 1.2239328201539539, "percentage": 24.44, "elapsed_time": "2:01:41", "remaining_time": "6:16:07"}
	{"current_steps": 120, "total_steps": 450, "loss": 0.2316, "lr": 1.83972518410775e-05, "epoch": 1.3358992302309307, "percentage": 26.67, "elapsed_time": "2:12:20", "remaining_time": "6:03:57"}
	{"current_steps": 130, "total_steps": 450, "loss": 0.2216, "lr": 1.795120796494848e-05, "epoch": 1.4478656403079078, "percentage": 28.89, "elapsed_time": "2:22:57", "remaining_time": "5:51:54"}
	{"current_steps": 140, "total_steps": 450, "loss": 0.2155, "lr": 1.7457344588544018e-05, "epoch": 1.5598320503848844, "percentage": 31.11, "elapsed_time": "2:33:35", "remaining_time": "5:40:06"}
	{"current_steps": 150, "total_steps": 450, "loss": 0.2122, "lr": 1.691863186431996e-05, "epoch": 1.6717984604618614, "percentage": 33.33, "elapsed_time": "2:44:14", "remaining_time": "5:28:28"}
	{"current_steps": 160, "total_steps": 450, "loss": 0.2059, "lr": 1.63383096739871e-05, "epoch": 1.7837648705388385, "percentage": 35.56, "elapsed_time": "2:54:50", "remaining_time": "5:16:54"}
	{"current_steps": 170, "total_steps": 450, "loss": 0.2034, "lr": 1.5719868143481385e-05, "epoch": 1.8957312806158153, "percentage": 37.78, "elapsed_time": "3:05:28", "remaining_time": "5:05:29"}
	{"current_steps": 180, "total_steps": 450, "loss": 0.1995, "lr": 1.5067026652935823e-05, "epoch": 2.0, "percentage": 40.0, "elapsed_time": "3:15:22", "remaining_time": "4:53:03"}
	{"current_steps": 190, "total_steps": 450, "loss": 0.1764, "lr": 1.4383711467890776e-05, "epoch": 2.111966410076977, "percentage": 42.22, "elapsed_time": "3:26:02", "remaining_time": "4:41:56"}
	{"current_steps": 200, "total_steps": 450, "loss": 0.1713, "lr": 1.3674032126270982e-05, "epoch": 2.2239328201539537, "percentage": 44.44, "elapsed_time": "3:36:41", "remaining_time": "4:30:52"}
	{"current_steps": 200, "total_steps": 450, "eval_loss": 0.18000730872154236, "epoch": 2.2239328201539537, "percentage": 44.44, "elapsed_time": "3:42:18", "remaining_time": "4:37:53"}
	{"current_steps": 210, "total_steps": 450, "loss": 0.1671, "lr": 1.2942256723140951e-05, "epoch": 2.3358992302309307, "percentage": 46.67, "elapsed_time": "3:53:31", "remaining_time": "4:26:53"}
	{"current_steps": 220, "total_steps": 450, "loss": 0.1703, "lr": 1.2192786241879033e-05, "epoch": 2.4478656403079078, "percentage": 48.89, "elapsed_time": "4:04:10", "remaining_time": "4:15:16"}
	{"current_steps": 230, "total_steps": 450, "loss": 0.1642, "lr": 1.1430128086145542e-05, "epoch": 2.5598320503848844, "percentage": 51.11, "elapsed_time": "4:14:49", "remaining_time": "4:03:45"}
	{"current_steps": 240, "total_steps": 450, "loss": 0.1553, "lr": 1.0658868971826785e-05, "epoch": 2.6717984604618614, "percentage": 53.33, "elapsed_time": "4:25:28", "remaining_time": "3:52:17"}
	{"current_steps": 250, "total_steps": 450, "loss": 0.1616, "lr": 9.883647341986032e-06, "epoch": 2.7837648705388385, "percentage": 55.56, "elapsed_time": "4:36:07", "remaining_time": "3:40:53"}
	{"current_steps": 260, "total_steps": 450, "loss": 0.157, "lr": 9.109125470721141e-06, "epoch": 2.8957312806158155, "percentage": 57.78, "elapsed_time": "4:46:45", "remaining_time": "3:29:33"}
	{"current_steps": 270, "total_steps": 450, "loss": 0.1528, "lr": 8.339961423699563e-06, "epoch": 3.0, "percentage": 60.0, "elapsed_time": "4:56:41", "remaining_time": "3:17:47"}
	{"current_steps": 280, "total_steps": 450, "loss": 0.1249, "lr": 7.580781044003324e-06, "epoch": 3.111966410076977, "percentage": 62.22, "elapsed_time": "5:07:25", "remaining_time": "3:06:39"}
	{"current_steps": 290, "total_steps": 450, "loss": 0.1168, "lr": 6.836150131764434e-06, "epoch": 3.2239328201539537, "percentage": 64.44, "elapsed_time": "5:18:07", "remaining_time": "2:55:31"}
	{"current_steps": 300, "total_steps": 450, "loss": 0.1178, "lr": 6.110546984905661e-06, "epoch": 3.3358992302309307, "percentage": 66.67, "elapsed_time": "5:28:51", "remaining_time": "2:44:25"}
	{"current_steps": 300, "total_steps": 450, "eval_loss": 0.15590140223503113, "epoch": 3.3358992302309307, "percentage": 66.67, "elapsed_time": "5:34:27", "remaining_time": "2:47:13"}
	{"current_steps": 310, "total_steps": 450, "loss": 0.1208, "lr": 5.4083354661298816e-06, "epoch": 3.4478656403079078, "percentage": 68.89, "elapsed_time": "5:45:41", "remaining_time": "2:36:07"}
	{"current_steps": 320, "total_steps": 450, "loss": 0.1144, "lr": 4.733738758136327e-06, "epoch": 3.5598320503848844, "percentage": 71.11, "elapsed_time": "5:56:22", "remaining_time": "2:24:46"}
	{"current_steps": 330, "total_steps": 450, "loss": 0.1156, "lr": 4.090813964902889e-06, "epoch": 3.6717984604618614, "percentage": 73.33, "elapsed_time": "6:07:04", "remaining_time": "2:13:28"}
	{"current_steps": 340, "total_steps": 450, "loss": 0.1132, "lr": 3.483427711785449e-06, "epoch": 3.7837648705388385, "percentage": 75.56, "elapsed_time": "6:17:45", "remaining_time": "2:02:12"}
	{"current_steps": 350, "total_steps": 450, "loss": 0.1116, "lr": 2.9152328911780027e-06, "epoch": 3.8957312806158155, "percentage": 77.78, "elapsed_time": "6:28:28", "remaining_time": "1:50:59"}
	{"current_steps": 360, "total_steps": 450, "loss": 0.1115, "lr": 2.3896466935879957e-06, "epoch": 4.0, "percentage": 80.0, "elapsed_time": "6:38:27", "remaining_time": "1:39:36"}
	{"current_steps": 370, "total_steps": 450, "loss": 0.0792, "lr": 1.9098300562505266e-06, "epoch": 4.111966410076977, "percentage": 82.22, "elapsed_time": "6:49:12", "remaining_time": "1:28:28"}
	{"current_steps": 380, "total_steps": 450, "loss": 0.0739, "lr": 1.4786686528798878e-06, "epoch": 4.223932820153954, "percentage": 84.44, "elapsed_time": "6:59:56", "remaining_time": "1:17:21"}
	{"current_steps": 390, "total_steps": 450, "loss": 0.0777, "lr": 1.0987555388883042e-06, "epoch": 4.335899230230931, "percentage": 86.67, "elapsed_time": "7:10:39", "remaining_time": "1:06:15"}
	{"current_steps": 400, "total_steps": 450, "loss": 0.0765, "lr": 7.723755564455771e-07, "epoch": 4.447865640307907, "percentage": 88.89, "elapsed_time": "7:21:23", "remaining_time": "0:55:10"}
	{"current_steps": 400, "total_steps": 450, "eval_loss": 0.16239598393440247, "epoch": 4.447865640307907, "percentage": 88.89, "elapsed_time": "7:26:59", "remaining_time": "0:55:52"}
	{"current_steps": 410, "total_steps": 450, "loss": 0.0784, "lr": 5.014915931694253e-07, "epoch": 4.559832050384885, "percentage": 91.11, "elapsed_time": "7:38:19", "remaining_time": "0:44:42"}
	{"current_steps": 420, "total_steps": 450, "loss": 0.0774, "lr": 2.877327770883964e-07, "epoch": 4.671798460461861, "percentage": 93.33, "elapsed_time": "7:49:05", "remaining_time": "0:33:30"}
	{"current_steps": 430, "total_steps": 450, "loss": 0.0736, "lr": 1.323846788742078e-07, "epoch": 4.783764870538838, "percentage": 95.56, "elapsed_time": "7:59:49", "remaining_time": "0:22:19"}
	{"current_steps": 440, "total_steps": 450, "loss": 0.0748, "lr": 3.6381580268463056e-08, "epoch": 4.8957312806158155, "percentage": 97.78, "elapsed_time": "8:10:35", "remaining_time": "0:11:08"}
	{"current_steps": 450, "total_steps": 450, "loss": 0.0727, "lr": 3.008552023242572e-10, "epoch": 5.0, "percentage": 100.0, "elapsed_time": "8:20:37", "remaining_time": "0:00:00"}
	{"current_steps": 450, "total_steps": 450, "epoch": 5.0, "percentage": 100.0, "elapsed_time": "8:21:09", "remaining_time": "0:00:00"}