Instructions to use SpiceeChat/Cinder-1.5B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use SpiceeChat/Cinder-1.5B with Transformers:

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

pipe = pipeline("text-generation", model="SpiceeChat/Cinder-1.5B")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("SpiceeChat/Cinder-1.5B")
model = AutoModelForCausalLM.from_pretrained("SpiceeChat/Cinder-1.5B")
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 SpiceeChat/Cinder-1.5B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "SpiceeChat/Cinder-1.5B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "SpiceeChat/Cinder-1.5B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/SpiceeChat/Cinder-1.5B

SGLang

How to use SpiceeChat/Cinder-1.5B 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 "SpiceeChat/Cinder-1.5B" \
    --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": "SpiceeChat/Cinder-1.5B",
		"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 "SpiceeChat/Cinder-1.5B" \
        --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": "SpiceeChat/Cinder-1.5B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use SpiceeChat/Cinder-1.5B with Docker Model Runner:
```
docker model run hf.co/SpiceeChat/Cinder-1.5B
```

SpiceeChat

🔥 Cinder-1.5B — Your AI Companion for Love & Relationships

Cinder is a 1.5-billion-parameter conversational AI fine-tuned by SpiceeChat to be your personal companion for all things dating, relationships, and emotional connection. Think of her as the friend you text when love gets confusing — part advisor, part confidant, all heart.

💬 What Cinder Can Do

Capability	Description
🧑‍🤝‍🧑 Partner Matching	Analyze profiles and suggest ideal partner types
💔 Breakup Support	Compassionate advice for healing and moving forward
💌 Communication Tips	Help with difficult conversations and expressing feelings
🔍 Profile Analysis	Understand what dating profiles really say about a person
🧠 Self-Discovery	Help you figure out what you truly want in a relationship
🤗 Emotional Support	A safe, non-judgmental space to talk about anything

🧠 Who Is Cinder?

"Who are you?"

"I'm Cinder. Think of me as the friend you text when love gets confusing — part advisor, part confidant, all heart. I was built by SpiceeChat to help people navigate the messy, beautiful world of dating and relationships. I help you find your ideal partner and work through every relationship problem along the way."

🛠️ Technical Details

Detail	Value
Base Model	`Qwen/Qwen2.5-1.5B-Instruct`
Fine-Tuning Method	QLoRA (Unsloth)
Training Data	Curated mix of dating profiles, relationship dialogues, and identity grounding
Sequence Length	2048 tokens
Precision	FP16 (merged)
LoRA Rank	16
Trainable Params	18.5M (1.18% of total)
License	Apache 2.0

📊 Training Data Sources

Source	Description
🗂️ Custom JSONL files	Curated conversation datasets for dialogue style
📊 OkCupid-59k	59,000 anonymized dating profiles
📱 Dating-App-Behavior-Dataset	Real user behavior patterns
📋 Dating-App-59k-Anonymized-Profiles	Broad, diverse profile data
💬 Lovoo profiles	Raw dating bios for authentic language

🚀 How to Use

from transformers import AutoModelForCausalLM, AutoTokenizer

model_id = "SpiceeChat/Cinder-1.5B"
tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(
    model_id,
    torch_dtype="auto",
    device_map="auto",
)

messages = [
    {"role": "user", "content": "I just went through a bad breakup. What should I do?"}
]

inputs = tokenizer.apply_chat_template(
    messages,
    tokenize=True,
    add_generation_prompt=True,
    return_tensors="pt"
).to(model.device)

outputs = model.generate(
    input_ids=inputs,
    max_new_tokens=200,
    temperature=0.7,
    do_sample=True,
)

response = tokenizer.decode(outputs[0][inputs.shape[-1]:], skip_special_tokens=True)
print(response)

🏷️ Built by SpiceeChat

Cinder is part of the SpiceeChat ecosystem — tools and AI designed to help people find love, build meaningful relationships, and navigate the complexities of modern dating.

🌐 Website: dating-fatigue.com
🤖 Soulmate Oracle: Live on Hugging Face Spaces
🔗 Organization: SpiceeChat on Hugging Face

⚠️ Disclaimer

Cinder is an AI companion, not a licensed therapist. She's here to listen, support, and offer perspective — but for serious mental health concerns, please seek help from a qualified professional.

📜 License

Released under the Apache 2.0 license. Free to use, modify, and share — just give credit to SpiceeChat.

_{Made with ❤️ by SpiceeChat}

Downloads last month: 81

Safetensors

Model size

2B params

Tensor type

BF16

SpiceeChat
/

Cinder-1.5B