Instructions to use John6666/llama-joycaption-alpha-two-hf-llava-nf4 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use John6666/llama-joycaption-alpha-two-hf-llava-nf4 with Transformers:

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

pipe = pipeline("image-text-to-text", model="John6666/llama-joycaption-alpha-two-hf-llava-nf4")
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("John6666/llama-joycaption-alpha-two-hf-llava-nf4")
model = AutoModelForMultimodalLM.from_pretrained("John6666/llama-joycaption-alpha-two-hf-llava-nf4", 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 John6666/llama-joycaption-alpha-two-hf-llava-nf4 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "John6666/llama-joycaption-alpha-two-hf-llava-nf4"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "John6666/llama-joycaption-alpha-two-hf-llava-nf4",
		"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/John6666/llama-joycaption-alpha-two-hf-llava-nf4

SGLang

How to use John6666/llama-joycaption-alpha-two-hf-llava-nf4 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 "John6666/llama-joycaption-alpha-two-hf-llava-nf4" \
    --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": "John6666/llama-joycaption-alpha-two-hf-llava-nf4",
		"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 "John6666/llama-joycaption-alpha-two-hf-llava-nf4" \
        --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": "John6666/llama-joycaption-alpha-two-hf-llava-nf4",
		"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 John6666/llama-joycaption-alpha-two-hf-llava-nf4 with Docker Model Runner:
```
docker model run hf.co/John6666/llama-joycaption-alpha-two-hf-llava-nf4
```

John6666 commited on Jan 29, 2025

Commit

ad32eb7

verified ·

1 Parent(s): ef9cfeb

Upload 3 files

Browse files

Files changed (1) hide show

scripts/README.md +22 -17

scripts/README.md CHANGED Viewed

@@ -14,23 +14,24 @@ This command will caption all the `.jpg` images in the specified directory using
 ## Command-Line Arguments
-**Note**: You must specify either `--glob` or `--filelist` to provide images, and either `--prompt` or `--prompt-file` to provide a prompt for caption generation.
-| Argument           | Description                                                | Default                                                                                                                 |
-| ------------------ | ---------------------------------------------------------- | ----------------------------------------------------------------------------------------------------------------------- |
-| `--glob`           | Glob pattern to find images                                | N/A                                                                                                                     |
-| `--filelist`       | File containing a list of images                           | N/A                                                                                                                     |
-| `--prompt`         | Prompt to use for caption generation                       | N/A                                                                                                                     |
-| `--prompt-file`    | JSON file containing prompts                               | N/A                                                                                                                     |
-| `--batch-size`     | Batch size for image processing                            | 1                                                                                                                       |
-| `--greedy`         | Use greedy decoding instead of sampling                    | False                                                                                                                   |
-| `--temperature`    | Sampling temperature (used when not using greedy decoding) | 0.6                                                                                                                     |
-| `--top-p`          | Top-p sampling value (nucleus sampling)                    | 0.9                                                                                                                     |
-| `--top-k`          | Top-k sampling value                                       | None                                                                                                                    |
-| `--max-new-tokens` | Maximum length of the generated caption (in tokens)        | 256                                                                                                                     |
-| `--num-workers`    | Number of workers loading images in parallel               | 4                                                                                                                       |
-| `--model`          | Pre-trained model to use                                   | [fancyfeast/llama-joycaption-alpha-two-hf-llava](https://huggingface.co/fancyfeast/llama-joycaption-alpha-two-hf-llava) |
 ### Examples
@@ -40,6 +41,10 @@ This command will caption all the `.jpg` images in the specified directory using
    ```sh
    ./batch-caption.py --glob "images/*.png" --prompt "Write a descriptive caption for this image in a formal tone."
    ```
 2. **Use a JSON file for prompts**

 ## Command-Line Arguments
+**Note**: You must specify either `--glob` or `--filelist` or `--input` to provide images, and either `--prompt` or `--prompt-file` to provide a prompt for caption generation.
+| Argument           | Description                                                | Default                                                                                                                     |
+| ------------------ | ---------------------------------------------------------- | --------------------------------------------------------------------------------------------------------------------------- |
+| `--input`          | Input images                                               | N/A                                                                                                                         |
+| `--glob`           | Glob pattern to find images                                | N/A                                                                                                                         |
+| `--filelist`       | File containing a list of images                           | N/A                                                                                                                         |
+| `--prompt`         | Prompt to use for caption generation                       | N/A                                                                                                                         |
+| `--prompt-file`    | JSON file containing prompts                               | N/A                                                                                                                         |
+| `--batch-size`     | Batch size for image processing                            | 1                                                                                                                           |
+| `--greedy`         | Use greedy decoding instead of sampling                    | False                                                                                                                       |
+| `--temperature`    | Sampling temperature (used when not using greedy decoding) | 0.6                                                                                                                         |
+| `--top-p`          | Top-p sampling value (nucleus sampling)                    | 0.9                                                                                                                         |
+| `--top-k`          | Top-k sampling value                                       | None                                                                                                                        |
+| `--max-new-tokens` | Maximum length of the generated caption (in tokens)        | 256                                                                                                                         |
+| `--num-workers`    | Number of workers loading images in parallel               | 4                                                                                                                           |
+| `--model`          | Pre-trained model to use                                   | [John6666/llama-joycaption-alpha-two-hf-llava-nf4](https://huggingface.co/John6666/llama-joycaption-alpha-two-hf-llava-nf4) |
+| `--bf16`           | Load model on torch.bfloat16                               | False                                                                                                                       |
 ### Examples
    ```sh
    ./batch-caption.py --glob "images/*.png" --prompt "Write a descriptive caption for this image in a formal tone."
    ```
+   or
+   ```sh
+   ./batch-caption.py --input "images/dog.png" --prompt "Write a descriptive caption for this image in a formal tone."
+   ```
 2. **Use a JSON file for prompts**