Instructions to use nex-agi/Nex-N2-Pro with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use nex-agi/Nex-N2-Pro with Transformers:

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

pipe = pipeline("text-generation", model="nex-agi/Nex-N2-Pro")
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("nex-agi/Nex-N2-Pro")
model = AutoModelForMultimodalLM.from_pretrained("nex-agi/Nex-N2-Pro", 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 nex-agi/Nex-N2-Pro with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "nex-agi/Nex-N2-Pro"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "nex-agi/Nex-N2-Pro",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/nex-agi/Nex-N2-Pro

SGLang

How to use nex-agi/Nex-N2-Pro 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 "nex-agi/Nex-N2-Pro" \
    --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": "nex-agi/Nex-N2-Pro",
		"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 "nex-agi/Nex-N2-Pro" \
        --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": "nex-agi/Nex-N2-Pro",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use nex-agi/Nex-N2-Pro with Docker Model Runner:
```
docker model run hf.co/nex-agi/Nex-N2-Pro
```

Next Generation of Nex-N2-Pro?

#17

by wake6 - opened 2 days ago

Discussion

wake6

2 days ago

•

edited 2 days ago

Next Generation of Nex-N2-Pro? Product / Strategy / Architect Thinking

First of all, thank you for Nex-N2-Pro and the work behind it. It has been one of the most useful open models for us, especially because its Agentic Thinking feels broader than just coding or tool execution.

Do you have any plan or timeline for the next generation of Nex-N2-Pro?

Many recent post-trained models—Ornith-1.5, DeepSeek-V4-Flash, Qwen3.8, GLM, etc.—are becoming extremely strong as Agent Engineers: coding, tool use, harness execution, debugging, repository work, and task completion.

Skills can scaffold a process, but they cannot fully substitute for the model's own knowledge, abstraction, judgment, problem formulation, and ability to construct a rich solution space.

However, we don't think Skills completely replace the underlying model capabilities. A Skill can tell a model what questions to ask, but the model still needs enough general/domain knowledge, judgment, abstraction, and open-ended reasoning to discover possibilities that were not explicitly encoded in the Skill.

In real product development, we increasingly need a “main brain” with Product / Strategy / Architect thinking:

challenge why a feature should exist before deciding how to implement it;
identify the real user/business problem instead of accepting a proposed solution too early;
explore fundamentally different product and architecture directions;
reason about business, domain, information, and system boundaries;
make difficult trade-offs where there is no single verifiable answer.

Capabilities such as requirement analysis, product design, business architecture, information architecture, system architecture, strategic analysis, and complex decision-making are much harder to train and benchmark than coding because they often lack objective rewards.

Coding can increasingly be delegated to specialized models. What is harder to find is a strong open model that can serve as the product / architect / decision-making main brain, with Skills acting as augmentation rather than compensation for missing core capabilities.

Is strengthening this kind of open-ended knowledge work—while preserving Nex-N2-Pro's strong general knowledge and Agentic Thinking—a direction you are considering for the next Nex generation?

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