Instructions to use JCabs/SunnyTeacher16Bv2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use JCabs/SunnyTeacher16Bv2 with Transformers:

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

pipe = pipeline("text-generation", model="JCabs/SunnyTeacher16Bv2")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("JCabs/SunnyTeacher16Bv2")
model = AutoModelForCausalLM.from_pretrained("JCabs/SunnyTeacher16Bv2")
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 JCabs/SunnyTeacher16Bv2 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "JCabs/SunnyTeacher16Bv2"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "JCabs/SunnyTeacher16Bv2",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/JCabs/SunnyTeacher16Bv2

SGLang

How to use JCabs/SunnyTeacher16Bv2 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 "JCabs/SunnyTeacher16Bv2" \
    --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": "JCabs/SunnyTeacher16Bv2",
		"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 "JCabs/SunnyTeacher16Bv2" \
        --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": "JCabs/SunnyTeacher16Bv2",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Unsloth Studio new

How to use JCabs/SunnyTeacher16Bv2 with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for JCabs/SunnyTeacher16Bv2 to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for JCabs/SunnyTeacher16Bv2 to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for JCabs/SunnyTeacher16Bv2 to start chatting

Load model with FastModel

pip install unsloth
from unsloth import FastModel
model, tokenizer = FastModel.from_pretrained(
    model_name="JCabs/SunnyTeacher16Bv2",
    max_seq_length=2048,
)

Docker Model Runner
How to use JCabs/SunnyTeacher16Bv2 with Docker Model Runner:
```
docker model run hf.co/JCabs/SunnyTeacher16Bv2
```

SunnyTeacher-8Bv2

SunnyTeacher-8Bv2 is a child-safe educational assistant model fine-tuned from Qwen3-8B.

Features

STEM teaching for ages 3–17
Step-by-step explanations
Child-safe responses
Emotional support guidance
Educational reasoning
Creativity and imagination support
-v2 added Emotional support guidance.

Training

Fine-tuned using Unsloth LoRA training on curated educational and child-safe datasets.

Intended Use

Educational assistance, tutoring, STEM learning, and safe child interaction. SunnyTeacher-16B is intended to help children and students with homework, STEM learning, and general educational questions by providing step-by-step explanations instead of only giving direct answers. The goal is to help learners understand the process, build confidence, and focus on building critical thinking in an age-appropriate way.

Base Model

Qwen3-8B

License

Apache-2.0

Training Framework

SunnyTeacher-8B was trained using:

PyTorch
Hugging Face Transformers
Unsloth
PEFT (LoRA fine-tuning)
TRL (Supervised Fine-Tuning)
Safetensors

Hardware

Training performed on:

Offline
NVIDIA GeForce RTX 4090
CUDA 12.8

Training Method

The model was fine-tuned using supervised instruction tuning on curated child-safe educational datasets focused on: - STEM education - Emotional safety - Age-appropriate explanations - Step-by-step reasoning - Child-safe conversational behavior

Disclaimer

SunnyTeacher-8Bv2 is currently a work in progress (WIP) research and educational model. While significant effort has been made to encourage safe, age-appropriate, and educational responses, the model may still occasionally generate incorrect, inappropriate, incomplete, or unsafe outputs. Parents, guardians, teachers, and developers should review responses before relying on them for child use cases. If you encounter a response that appears unsafe, inappropriate, or inaccurate for children, please provide feedback or report the issue so the model can continue improving.

Downloads last month: 38

Safetensors

Model size

8B params

Tensor type

BF16

Model tree for JCabs/SunnyTeacher16Bv2

Base model

Qwen/Qwen3-8B-Base

Finetuned

Qwen/Qwen3-8B

Adapter

(1263)

this model