Instructions to use adowu/astral-128k with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use adowu/astral-128k with Transformers:

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

pipe = pipeline("text-generation", model="adowu/astral-128k")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("adowu/astral-128k")
model = AutoModelForCausalLM.from_pretrained("adowu/astral-128k")
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 adowu/astral-128k with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "adowu/astral-128k"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "adowu/astral-128k",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/adowu/astral-128k

SGLang

How to use adowu/astral-128k 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 "adowu/astral-128k" \
    --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": "adowu/astral-128k",
		"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 "adowu/astral-128k" \
        --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": "adowu/astral-128k",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use adowu/astral-128k with Docker Model Runner:
```
docker model run hf.co/adowu/astral-128k
```

ASTRAL-128K

Model Description

The adowu/astral-128k model is a state-of-the-art language model based on the MistralForCausalLM architecture, designed for causal text generation. This means the model generates text sequentially, where each subsequent word is chosen based on the previous words in the sequence. Below is an overview of the key features and configuration of this model.

Architectural Highlights

The model is built on a foundation designed to understand and generate language by sequentially predicting the next piece of text based on the preceding context. Its architecture allows for deep understanding of language nuances, enabling the generation of text that is not only relevant but also rich in variety.

Capabilities

With its advanced architecture, the model can handle a wide range of language generation tasks. It is capable of generating long passages of text that maintain coherence and relevance throughout, making it suitable for tasks such as content creation, conversation generation, and more complex applications like summarization and translation in context-rich scenarios.

Performance and Efficiency

The model is optimized for both performance and efficiency, ensuring that it can process and generate text quickly while maintaining a high level of accuracy and coherence in the output. This balance makes it an excellent tool for developers and researchers looking for a reliable model for generating high-quality text.

Application Scope

Due to its versatile nature, the model finds applications across various domains, including but not limited to, automated storytelling, chatbots, and assistance systems, content generation for websites, and aiding in the creation of written content for educational and informational purposes.

In summary, the adowu/astral-128k model represents a significant advancement in the field of natural language processing, offering a powerful tool for generating human-like text across a wide array of applications.

Developed by: aww
Model type: Mistral

Downloads last month: 12

Safetensors

Model size

4B params

Tensor type

BF16