Instructions to use demo-at-wsld-8/blip-at-wsld-fashion200k with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use demo-at-wsld-8/blip-at-wsld-fashion200k with Transformers:

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

pipe = pipeline("image-text-to-text", model="demo-at-wsld-8/blip-at-wsld-fashion200k")

# Load model directly
from transformers import AutoProcessor, AutoModelForMultimodalLM

processor = AutoProcessor.from_pretrained("demo-at-wsld-8/blip-at-wsld-fashion200k")
model = AutoModelForMultimodalLM.from_pretrained("demo-at-wsld-8/blip-at-wsld-fashion200k", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use demo-at-wsld-8/blip-at-wsld-fashion200k with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "demo-at-wsld-8/blip-at-wsld-fashion200k"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "demo-at-wsld-8/blip-at-wsld-fashion200k",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/demo-at-wsld-8/blip-at-wsld-fashion200k

SGLang

How to use demo-at-wsld-8/blip-at-wsld-fashion200k 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 "demo-at-wsld-8/blip-at-wsld-fashion200k" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "demo-at-wsld-8/blip-at-wsld-fashion200k",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

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 "demo-at-wsld-8/blip-at-wsld-fashion200k" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "demo-at-wsld-8/blip-at-wsld-fashion200k",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use demo-at-wsld-8/blip-at-wsld-fashion200k with Docker Model Runner:
```
docker model run hf.co/demo-at-wsld-8/blip-at-wsld-fashion200k
```

BLIP-base distilled (Attention Transfer + WSLD) — Fashion-200k captioning

Student model distill từ Salesforce/blip-image-captioning-large (Teacher) sang Salesforce/blip-image-captioning-base đã prune decoder (12 → 4 tầng), train trên Fashion-200k bằng combined loss:

Attention Transfer (AT) ở vision encoder — so khớp attention map giữa Student/Teacher.
WSLD (Weighted Soft Label Distillation) ở decoder — soft loss có trọng số theo token.
Cross-entropy với ground-truth caption.

Kiến trúc

Tham số: 148.4M | Kích thước: 283.0MB
Decoder giữ lại tầng: [0, 3, 6, 9]

Kết quả trên test set

{ "BLEU-4": 0.2458812827818811, "BLEU-1": 0.6297635645404903, "ROUGE-1": 0.5509984919361472, "ROUGE-2": 0.33081984865850167, "ROUGE-L": 0.4506778285021464, "METEOR": 0.40850410599132086, "CIDEr": 0.6211192626331922 }

Cách dùng

from huggingface_hub import hf_hub_download
from transformers import BlipForConditionalGeneration, BlipProcessor
import torch, json

repo_id = "demo-at-wsld-8/blip-at-wsld-fashion200k"
cfg = json.load(open(hf_hub_download(repo_id, "deploy_config.json"), encoding="utf-8"))
ckpt = hf_hub_download(repo_id, "student_at_wsld.pth")
# xem code dung lai kien truc (prune_decoder_layers, fix_decoder_layer_idx) trong app.py cua Space demo

Code training đầy đủ

Xem notebook Kaggle: https://www.kaggle.com/code/anhvunguyenquynh/at-wsld

Downloads last month: 60

Safetensors

Model size

0.1B params

Tensor type

F16

Inference Providers NEW

Image-Text-to-Text

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

demo-at-wsld-8
/

blip-at-wsld-fashion200k

BLIP-base distilled (Attention Transfer + WSLD) — Fashion-200k captioning

Kiến trúc

Kết quả trên test set

Cách dùng

Code training đầy đủ

Space using demo-at-wsld-8/blip-at-wsld-fashion200k 1