Instructions to use bluuluu/Qwen2-VL-2B-Instruct-SFT with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use bluuluu/Qwen2-VL-2B-Instruct-SFT with Transformers:

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

pipe = pipeline("image-text-to-text", model="bluuluu/Qwen2-VL-2B-Instruct-SFT")
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)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoProcessor, AutoModelForMultimodalLM

processor = AutoProcessor.from_pretrained("bluuluu/Qwen2-VL-2B-Instruct-SFT")
model = AutoModelForMultimodalLM.from_pretrained("bluuluu/Qwen2-VL-2B-Instruct-SFT", 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=256)
print(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use bluuluu/Qwen2-VL-2B-Instruct-SFT with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "bluuluu/Qwen2-VL-2B-Instruct-SFT"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "bluuluu/Qwen2-VL-2B-Instruct-SFT",
		"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/bluuluu/Qwen2-VL-2B-Instruct-SFT

SGLang

How to use bluuluu/Qwen2-VL-2B-Instruct-SFT 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 "bluuluu/Qwen2-VL-2B-Instruct-SFT" \
    --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": "bluuluu/Qwen2-VL-2B-Instruct-SFT",
		"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 "bluuluu/Qwen2-VL-2B-Instruct-SFT" \
        --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": "bluuluu/Qwen2-VL-2B-Instruct-SFT",
		"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 bluuluu/Qwen2-VL-2B-Instruct-SFT with Docker Model Runner:
```
docker model run hf.co/bluuluu/Qwen2-VL-2B-Instruct-SFT
```

bluuluu commited on Jun 2, 2025

Commit

da9c6e2

verified ·

1 Parent(s): 7c48622

Model save

Browse files

Files changed (4) hide show

README.md +2 -5
all_results.json +7 -7
train_results.json +7 -7
trainer_state.json +0 -0

README.md CHANGED Viewed

@@ -1,11 +1,8 @@
 ---
-base_model: Qwen/Qwen2-VL-2B-Instruct
-datasets: MMInstruction/Clevr_CoGenT_TrainA_R1
 library_name: transformers
 model_name: Qwen2-VL-2B-Instruct-SFT
 tags:
 - generated_from_trainer
-- R1-V
 - trl
 - sft
 licence: license
@@ -13,7 +10,7 @@ licence: license
 # Model Card for Qwen2-VL-2B-Instruct-SFT
-This model is a fine-tuned version of [Qwen/Qwen2-VL-2B-Instruct](https://huggingface.co/Qwen/Qwen2-VL-2B-Instruct) on the [MMInstruction/Clevr_CoGenT_TrainA_R1](https://huggingface.co/datasets/MMInstruction/Clevr_CoGenT_TrainA_R1) dataset.
 It has been trained using [TRL](https://github.com/huggingface/trl).
 ## Quick start
@@ -29,7 +26,7 @@ print(output["generated_text"])
 ## Training procedure
-[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/2741919970-hustvl/huggingface/runs/kd2rv0e6)
 This model was trained with SFT.

 ---
 library_name: transformers
 model_name: Qwen2-VL-2B-Instruct-SFT
 tags:
 - generated_from_trainer
 - trl
 - sft
 licence: license
 # Model Card for Qwen2-VL-2B-Instruct-SFT
+This model is a fine-tuned version of [None](https://huggingface.co/None).
 It has been trained using [TRL](https://github.com/huggingface/trl).
 ## Quick start
 ## Training procedure
 This model was trained with SFT.

all_results.json CHANGED Viewed

@@ -1,9 +1,9 @@
 {
-    "epoch": 0.9998919736415686,
-    "total_flos": 2.0789981999005696e+16,
-    "train_loss": 0.0,
-    "train_runtime": 2.6266,
-    "train_samples": 40742,
-    "train_samples_per_second": 15511.09,
-    "train_steps_per_second": 242.135
 }

 {
+    "epoch": 1.9998919736415686,
+    "total_flos": 2.1613118366416896e+16,
+    "train_loss": 0.0653228780393709,
+    "train_runtime": 98246.8444,
+    "train_samples": 222165,
+    "train_samples_per_second": 4.523,
+    "train_steps_per_second": 0.071
 }

train_results.json CHANGED Viewed

@@ -1,9 +1,9 @@
 {
-    "epoch": 0.9998919736415686,
-    "total_flos": 2.0789981999005696e+16,
-    "train_loss": 0.0,
-    "train_runtime": 2.6266,
-    "train_samples": 40742,
-    "train_samples_per_second": 15511.09,
-    "train_steps_per_second": 242.135
 }

 {
+    "epoch": 1.9998919736415686,
+    "total_flos": 2.1613118366416896e+16,
+    "train_loss": 0.0653228780393709,
+    "train_runtime": 98246.8444,
+    "train_samples": 222165,
+    "train_samples_per_second": 4.523,
+    "train_steps_per_second": 0.071
 }

trainer_state.json CHANGED Viewed

The diff for this file is too large to render. See raw diff