Instructions to use HelpingAI/HAI-SER with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use HelpingAI/HAI-SER with Transformers:

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

pipe = pipeline("text-generation", model="HelpingAI/HAI-SER")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("HelpingAI/HAI-SER")
model = AutoModelForCausalLM.from_pretrained("HelpingAI/HAI-SER")
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 HelpingAI/HAI-SER with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/HelpingAI/HAI-SER

SGLang

How to use HelpingAI/HAI-SER 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 "HelpingAI/HAI-SER" \
    --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": "HelpingAI/HAI-SER",
		"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 "HelpingAI/HAI-SER" \
        --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": "HelpingAI/HAI-SER",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use HelpingAI/HAI-SER with Docker Model Runner:
```
docker model run hf.co/HelpingAI/HAI-SER
```

❤️ HAI-SER

[📜 License](https://helpingai.co/license) | [🌐 Website](https://helpingai.co)

🌟 About HAI-SER

HAI-SER is HelpingAI's revolutionary Structured Emotional Reasoning (SER) model, crafted to redefine the emotional intelligence of AI. Unlike traditional models, HAI-SER goes beyond words—it understands emotions, breaks down mental states, and offers real, empathetic insights for human-AI interaction. 🚀

💡 Core Features of HAI-SER

The Structured Emotional Reasoning (SER) framework is built upon these key pillars:

Emotional Vibe Check – Reads emotional energy from conversations 🎭
Mind-State Analysis – Understands thoughts, moods, and mental shifts 🧠
Root Cause Deep-Dive – Identifies why emotions arise 🔍
Impact Check – Evaluates how emotions affect real-life actions 💥
Safety Check – Prioritizes user well-being 🚨
Healing Game Plan – Offers structured support for growth & recovery 💪
Growth Potential – Helps users evolve emotionally 📈
How to Approach – Guides users in communication & self-awareness 🤝

🚀 Implementation

Load HAI-SER with Hugging Face Transformers

from transformers import AutoModelForCausalLM, AutoTokenizer

# Load HAI-SER
model = AutoModelForCausalLM.from_pretrained("HelpingAI/HAI-SER")
tokenizer = AutoTokenizer.from_pretrained("HelpingAI/HAI-SER")

# Example usage
chat = [
    {"role": "system", "content": "You are an emotionally intelligent AI assistant who always thinks step by step before responding."},
    {"role": "user", "content": "I feel really stressed out about my exams."}
]

inputs = tokenizer.apply_chat_template(
    chat,
    add_generation_prompt=True,
    return_tensors="pt"
)

outputs = model.generate(
    inputs,
    max_new_tokens=128,
    temperature=0.7,
    top_p=0.9,
)

print(tokenizer.decode(outputs[0], skip_special_tokens=True))

⚙️ Training Details

🏋️ Training Data

Trained on a curated dataset emphasizing emotional intelligence, human psychology, and nuanced conversation.
Includes dialogues from mental health scenarios, coaching sessions, and empathetic responses.

📌 Capabilities

Understands and analyzes emotions with high accuracy.
Provides tailored emotional insights instead of generic responses.
Capable of deep reasoning for emotional problem-solving.

⚠️ Limitations

Still evolving – may not always capture deep emotions perfectly.
Not a replacement for professional therapy – designed to support, not diagnose.
Best used with human moderation in sensitive situations.

📚 Citation

@misc{haiser2025,
  author = {HelpingAI Team},
  title = {HAI-SER: Structured Emotional Reasoning for Empathetic AI},
  year = {2025},
  publisher = {HelpingAI},
  journal = {HuggingFace},
  howpublished = {\url{https://huggingface.co/HelpingAI/HAI-SER}}
}

Created with ❤️ by HelpingAI

🌐 Website • 📜 License • 🤗 HuggingFace • 💬 Discord

Downloads last month: 16

Safetensors

Model size

14B params

Tensor type

F16

HelpingAI
/

HAI-SER