Instructions to use Lythri/Lythri-Tada-4B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Lythri/Lythri-Tada-4B with Transformers:

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

pipe = pipeline("text-generation", model="Lythri/Lythri-Tada-4B")
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("Lythri/Lythri-Tada-4B")
model = AutoModelForMultimodalLM.from_pretrained("Lythri/Lythri-Tada-4B", 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 Lythri/Lythri-Tada-4B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Lythri/Lythri-Tada-4B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Lythri/Lythri-Tada-4B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Lythri/Lythri-Tada-4B

SGLang

How to use Lythri/Lythri-Tada-4B 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 "Lythri/Lythri-Tada-4B" \
    --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": "Lythri/Lythri-Tada-4B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

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 "Lythri/Lythri-Tada-4B" \
        --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": "Lythri/Lythri-Tada-4B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use Lythri/Lythri-Tada-4B with Docker Model Runner:
```
docker model run hf.co/Lythri/Lythri-Tada-4B
```

Spike8086-HF commited on 2 days ago

Commit

035ec07

verified ·

1 Parent(s): 0ea0014

Update README.md

Browse files

Files changed (1) hide show

README.md +2 -2

README.md CHANGED Viewed

@@ -24,9 +24,9 @@ pipeline_tag: text-generation
 ## Introduction
-**Lythri-Tada-4B** is the instant-reply member of the Lythri family. It answers directly without a thinking phase, while keeping Lythri's focus on understanding how people feel and staying small enough to run locally on a laptop or phone.
-Compared with Lythri-4B-A2B, Tada brings two main upgrades:
 - **Stronger instruction following.** IFEval prompt-level strict accuracy rises from 26.43 to 39.19 (+12.76), and instruction-level strict accuracy reaches 49.88, so Tada follows formatting, length and style requests far more reliably.
 - **Better multi-turn conversations.** Tada keeps track of earlier context and user requests more consistently over long chats, and carries the emotional thread from turn to turn.

 ## Introduction
+**Lythri-Tada-4B** is part of the **Lythri 2.0** family. It answers directly without a thinking phase, while keeping Lythri's focus on understanding how people feel and staying small enough to run locally on a laptop or phone.
+Compared with the previous-generation Lythri-4B-A2B, Tada brings two main upgrades:
 - **Stronger instruction following.** IFEval prompt-level strict accuracy rises from 26.43 to 39.19 (+12.76), and instruction-level strict accuracy reaches 49.88, so Tada follows formatting, length and style requests far more reliably.
 - **Better multi-turn conversations.** Tada keeps track of earlier context and user requests more consistently over long chats, and carries the emotional thread from turn to turn.