Instructions to use dicta-il/dictalm2.0-instruct with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use dicta-il/dictalm2.0-instruct with Transformers:

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

pipe = pipeline("text-generation", model="dicta-il/dictalm2.0-instruct")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("dicta-il/dictalm2.0-instruct")
model = AutoModelForCausalLM.from_pretrained("dicta-il/dictalm2.0-instruct")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.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(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use dicta-il/dictalm2.0-instruct with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "dicta-il/dictalm2.0-instruct"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "dicta-il/dictalm2.0-instruct",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/dicta-il/dictalm2.0-instruct

SGLang

How to use dicta-il/dictalm2.0-instruct 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 "dicta-il/dictalm2.0-instruct" \
    --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": "dicta-il/dictalm2.0-instruct",
		"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 "dicta-il/dictalm2.0-instruct" \
        --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": "dicta-il/dictalm2.0-instruct",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use dicta-il/dictalm2.0-instruct with Docker Model Runner:
```
docker model run hf.co/dicta-il/dictalm2.0-instruct
```

Instruction prompting

by Tomor0720 - opened May 9, 2024

Discussion

Tomor0720

May 9, 2024

Does the instruction fine-tune includes mainly Hebrew text? It is better to prompt the model for instruction in Hebrew or in English?

Shaltiel

DICTA: The Israel Center for Text Analysis org May 12, 2024

The model is capable of understanding prompts in both English and Hebrew. Depending on the task, the model may respond better to instructions to one of the languages. Therefore, I recommend trying out the various prompts in the demo to see which produces the best results for your task consistently.

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment