Instructions to use Spravil/caption-via-translation-1_0B-ft with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Spravil/caption-via-translation-1_0B-ft with Transformers:

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

pipe = pipeline("image-text-to-text", model="Spravil/caption-via-translation-1_0B-ft", trust_remote_code=True)

# Load model directly
from transformers import AutoProcessor, AutoModelForImageTextToText

processor = AutoProcessor.from_pretrained("Spravil/caption-via-translation-1_0B-ft", trust_remote_code=True)
model = AutoModelForImageTextToText.from_pretrained("Spravil/caption-via-translation-1_0B-ft", trust_remote_code=True)

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use Spravil/caption-via-translation-1_0B-ft with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Spravil/caption-via-translation-1_0B-ft"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Spravil/caption-via-translation-1_0B-ft",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/Spravil/caption-via-translation-1_0B-ft

SGLang

How to use Spravil/caption-via-translation-1_0B-ft 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 "Spravil/caption-via-translation-1_0B-ft" \
    --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": "Spravil/caption-via-translation-1_0B-ft",
		"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 "Spravil/caption-via-translation-1_0B-ft" \
        --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": "Spravil/caption-via-translation-1_0B-ft",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use Spravil/caption-via-translation-1_0B-ft with Docker Model Runner:
```
docker model run hf.co/Spravil/caption-via-translation-1_0B-ft
```

Scaling Laws for Conditional Emergence of Multilingual Image Captioning via Generalization from Translation

1.0B Model (Fine-tuned)

The fine-tuned variant of the Pre-trained 1.0B Model is supervised on a curated mix of downstream tasks to broaden coverage and improve benchmark performance. Concretely, it is fine-tuned on multimodal translation and ambiguity/disambiguation as well as captioning across multiple caption styles, with missing languages filled in via MT augmentation. Downstream datasets used are: Multi30K, CoMMuTE, COCO Karpathy, XM3600, Image Paragraph, and DOCCI. The model supports the languages English, German, French, Spanish, Russian, and Chinese.

Getting Started

import requests
from PIL import Image
import torch
from transformers import AutoModelForCausalLM, AutoConfig, AutoProcessor, AutoTokenizer


device = "cuda:0" if torch.cuda.is_available() else "cpu"
torch_dtype = torch.float16 if torch.cuda.is_available() else torch.float32
model = AutoModelForCausalLM.from_pretrained("Spravil/caption-via-translation-1_0B-ft", torch_dtype=torch_dtype, trust_remote_code=True).to(device)
tokenizer = AutoTokenizer.from_pretrained(
    "google/gemma-2-2b",
    add_bos_token=True,
    add_eos_token=True,
    padding_side="right",
    truncation_side="right",
)
processor = AutoProcessor.from_pretrained("Spravil/caption-via-translation-1_0B-ft", trust_remote_code=True, new_tokenizer=tokenizer, use_encoder_tokenizer=True)
task = "<MORE_DETAILED_CAPTION>"
lang = "de"
prompt = f"<LANG_{lang.upper()}>{task}"
url = "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/transformers/tasks/car.jpg?download=true"
image = Image.open(requests.get(url, stream=True).raw).convert("RGB")
inputs = processor(prompt, images=image, return_tensors="pt").to(device, torch_dtype)
generated_ids = model.generate(
            **inputs,
            max_new_tokens=128,
            num_beams=4,
            do_sample=False,
            use_cache=False,
        )
generated_text = processor.batch_decode(generated_ids, skip_special_tokens=False)[0]
parsed_answer = processor.post_process_generation(generated_text, task=task, image_size=(image.width, image.height))
print(parsed_answer)

Bibtex

@inproceedings{spravil2026scaling,
  title={Scaling Laws for Conditional Emergence of Multilingual Image Captioning via Generalization from Translation},
  author={Spravil, Julian and Houben, Sebastian and Behnke, Sven},
  booktitle={Proceedings of the 40th AAAI Conference on Artificial Intelligence},
  year={2026}
}