Instructions to use anonseoul/qwen25vl7b-pgn-clean-ckpt5000 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use anonseoul/qwen25vl7b-pgn-clean-ckpt5000 with Transformers:

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

pipe = pipeline("image-text-to-text", model="anonseoul/qwen25vl7b-pgn-clean-ckpt5000")
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 AutoModelForMultimodalLM
model = AutoModelForMultimodalLM.from_pretrained("anonseoul/qwen25vl7b-pgn-clean-ckpt5000", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use anonseoul/qwen25vl7b-pgn-clean-ckpt5000 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "anonseoul/qwen25vl7b-pgn-clean-ckpt5000"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "anonseoul/qwen25vl7b-pgn-clean-ckpt5000",
		"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/anonseoul/qwen25vl7b-pgn-clean-ckpt5000

SGLang

How to use anonseoul/qwen25vl7b-pgn-clean-ckpt5000 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 "anonseoul/qwen25vl7b-pgn-clean-ckpt5000" \
    --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": "anonseoul/qwen25vl7b-pgn-clean-ckpt5000",
		"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 "anonseoul/qwen25vl7b-pgn-clean-ckpt5000" \
        --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": "anonseoul/qwen25vl7b-pgn-clean-ckpt5000",
		"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 anonseoul/qwen25vl7b-pgn-clean-ckpt5000 with Docker Model Runner:
```
docker model run hf.co/anonseoul/qwen25vl7b-pgn-clean-ckpt5000
```

anonseoul/qwen25vl7b-pgn-clean-ckpt5000

Artifact type: weights-only.

Source run: tvc_Qwen2_5-VL-7B-Instruct_v2_pgn_maximg672_bs4_ga4_h200x4_save500_max24465 W&B: https://wandb.ai/mir_lab/mm-pgn/runs/d1uc2qf9 Checkpoint step: 5000 Source path on launch pod: /home/irteam/data/junhyeok/cache_backup/tvc_Qwen2_5-VL-7B-Instruct_v2_pgn_maximg672_bs4_ga4_h200x4_save500_max24465/checkpoint-5000

Warning

Clean pre-break checkpoint from the original seed-42 run. The original run later showed a severe post-5000 copy-metric collapse, so this checkpoint is kept as the recovery source rather than as evidence that later steps were healthy.

Metrics At Checkpoint

Metric	Value
eval/loss	0.308446
eval/mean_copy_accuracy	0.995055
eval/mean_gen_accuracy	0.884841
eval/mean_token_accuracy	0.897334
eval/z_loss	0.000583555
eval/copy_logits_max	-6.47397
train/loss at step 4999	0.2573
train/mean_copy_accuracy at step 4999	0.995776
train/mean_gen_accuracy at step 4999	0.891306
train/mean_token_accuracy at step 4999	0.912032

File Contents

This repo excludes global_step*, rng_state*.pth, scheduler.pt, trainer_state.json, training_args.bin, and zero_to_fp32.py.

Downloads last month: 2

Safetensors

Model size

856k params

Tensor type

BF16

Model tree for anonseoul/qwen25vl7b-pgn-clean-ckpt5000

Base model

Qwen/Qwen2.5-VL-7B-Instruct

Finetuned

(1200)

this model