Instructions to use blueprint-ai/Blueprint-9B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use blueprint-ai/Blueprint-9B with Transformers:

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

pipe = pipeline("text-generation", model="blueprint-ai/Blueprint-9B")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("blueprint-ai/Blueprint-9B")
model = AutoModelForCausalLM.from_pretrained("blueprint-ai/Blueprint-9B")
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 blueprint-ai/Blueprint-9B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "blueprint-ai/Blueprint-9B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "blueprint-ai/Blueprint-9B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/blueprint-ai/Blueprint-9B

SGLang

How to use blueprint-ai/Blueprint-9B 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 "blueprint-ai/Blueprint-9B" \
    --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": "blueprint-ai/Blueprint-9B",
		"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 "blueprint-ai/Blueprint-9B" \
        --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": "blueprint-ai/Blueprint-9B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use blueprint-ai/Blueprint-9B with Docker Model Runner:
```
docker model run hf.co/blueprint-ai/Blueprint-9B
```

Blueprint-9B

Blueprint-9B is a specialized fine-tune of the Qwen 3.5-9B architecture, optimized for logical reasoning and interpreting unstructured technical instructions. It is designed to act as a project architect, prioritizing structural logic over simple syntax generation.

Performance Benchmarks

Metric	Blueprint-9B	Qwen 3.5 9B (Base)
Logic (GSM8K)	83.8%	81.0%
Code (HumanEval)	62.2%	81.7%
Scripts (MBPP)	62.0%	82.0%

Note: This model trades raw coding speed for higher reasoning accuracy, making it ideal for planning complex projects from informal notes.

###licensing & support Blueprint-9B is released under the Creative Commons Attribution-ShareAlike 3.0 Unported (CC BY-SA 3.0) license. Technical Support & Professional Inquiries: For support, please use the Discussions tab or contact: blueprintai.help1@gmail.com

###Official Credits & Legal Acknowledgments To ensure full compliance with open-source standards and respect for foundational work, we provide the following credits. This model is a derivative work distributed under the same license as the training data: Base Architecture: Developed by the Alibaba Cloud Qwen Team. We credit them for the high-performance Qwen 3.5-9B model foundation. Primary Dataset: Databricks, Inc. for the databricks-dolly-15k dataset (Copyright 2023 Databricks, Inc.). This model strictly adheres to the CC BY-SA 3.0 requirements as mandated by the Dolly dataset. Coding Data: Credits to the BigCode Project for the StarCoder instruct datasets and the m-a-p team for CodeFeedback. Instruction Tuning: Recognition to Tarun Sharma for the Alpaca-based Python instruction sets.

Implementation

from transformers import AutoModelForCausalLM, AutoTokenizer
import torch

model_id = "BlueprintLabs/Blueprint-9B"
tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(
    model_id, 
    device_map="auto", 
    torch_dtype=torch.bfloat16
)

Downloads last month: 137

Safetensors

Model size

9B params

Tensor type

BF16

Model tree for blueprint-ai/Blueprint-9B

Base model

Qwen/Qwen3.5-9B-Base

Finetuned

Qwen/Qwen3.5-9B

Finetuned

(232)

this model