Instructions to use ericflo/Llama-3.2-3B-COTv3 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use ericflo/Llama-3.2-3B-COTv3 with Transformers:

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

pipe = pipeline("text-generation", model="ericflo/Llama-3.2-3B-COTv3")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("ericflo/Llama-3.2-3B-COTv3")
model = AutoModelForCausalLM.from_pretrained("ericflo/Llama-3.2-3B-COTv3")
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]:]))

llama-cpp-python

How to use ericflo/Llama-3.2-3B-COTv3 with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="ericflo/Llama-3.2-3B-COTv3",
	filename="Llama-3.2-3B-COTv3-BF16.gguf",
)

llm.create_chat_completion(
	messages = [
		{
			"role": "user",
			"content": "What is the capital of France?"
		}
	]
)

Notebooks
Google Colab
Kaggle
Local Apps

llama.cpp

How to use ericflo/Llama-3.2-3B-COTv3 with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf ericflo/Llama-3.2-3B-COTv3:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf ericflo/Llama-3.2-3B-COTv3:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf ericflo/Llama-3.2-3B-COTv3:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf ericflo/Llama-3.2-3B-COTv3:Q4_K_M

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf ericflo/Llama-3.2-3B-COTv3:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf ericflo/Llama-3.2-3B-COTv3:Q4_K_M

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf ericflo/Llama-3.2-3B-COTv3:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf ericflo/Llama-3.2-3B-COTv3:Q4_K_M

Use Docker

docker model run hf.co/ericflo/Llama-3.2-3B-COTv3:Q4_K_M

LM Studio
Jan

vLLM

How to use ericflo/Llama-3.2-3B-COTv3 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "ericflo/Llama-3.2-3B-COTv3"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ericflo/Llama-3.2-3B-COTv3",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/ericflo/Llama-3.2-3B-COTv3:Q4_K_M

SGLang

How to use ericflo/Llama-3.2-3B-COTv3 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 "ericflo/Llama-3.2-3B-COTv3" \
    --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": "ericflo/Llama-3.2-3B-COTv3",
		"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 "ericflo/Llama-3.2-3B-COTv3" \
        --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": "ericflo/Llama-3.2-3B-COTv3",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Ollama
How to use ericflo/Llama-3.2-3B-COTv3 with Ollama:
```
ollama run hf.co/ericflo/Llama-3.2-3B-COTv3:Q4_K_M
```

Unsloth Studio new

How to use ericflo/Llama-3.2-3B-COTv3 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 ericflo/Llama-3.2-3B-COTv3 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 ericflo/Llama-3.2-3B-COTv3 to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for ericflo/Llama-3.2-3B-COTv3 to start chatting

Pi new

How to use ericflo/Llama-3.2-3B-COTv3 with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf ericflo/Llama-3.2-3B-COTv3:Q4_K_M

Configure the model in Pi

# Install Pi:
npm install -g @mariozechner/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "ericflo/Llama-3.2-3B-COTv3:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use ericflo/Llama-3.2-3B-COTv3 with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf ericflo/Llama-3.2-3B-COTv3:Q4_K_M

Configure Hermes

# Install Hermes:
curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash
hermes setup
# Point Hermes at the local server:
hermes config set model.provider custom
hermes config set model.base_url http://127.0.0.1:8080/v1
hermes config set model.default ericflo/Llama-3.2-3B-COTv3:Q4_K_M

Run Hermes

hermes

Docker Model Runner
How to use ericflo/Llama-3.2-3B-COTv3 with Docker Model Runner:
```
docker model run hf.co/ericflo/Llama-3.2-3B-COTv3:Q4_K_M
```

Lemonade

How to use ericflo/Llama-3.2-3B-COTv3 with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull ericflo/Llama-3.2-3B-COTv3:Q4_K_M

Run and chat with the model

lemonade run user.Llama-3.2-3B-COTv3-Q4_K_M

List all available models

lemonade list

Thought-Ranked Llama 3.2 3B v3.0

What's New in v3?

The major advancement in v3 is the integration of reinforcement learning to refine the model's outputs. Using OpenRLHF with REINFORCE and Gemini 1.5 Flash 8B as a judge, we've optimized the model to produce higher quality responses across various criteria including relevance, accuracy, clarity, style, and completeness.

This RL fine-tuning process used a sophisticated reward model that evaluates responses on a 0-99 scale, considering factors such as:

Intent fulfillment and practical utility
Factual accuracy and logical consistency
Clarity and understandability
Style and tone appropriateness
Completeness and detail sufficiency

How It Works

The model maintains the same powerful thought chain capabilities from v2.2, but with enhanced output quality. Here's an example:

<thoughts>
<thought>First, I should consider the moon's main effects on Earth</thought>
<thought>The moon controls our tides, so ocean patterns would change dramatically</thought>
<thought>Without the moon's gravitational pull, Earth's rotation would become unstable</thought>
<thought>This would lead to extreme climate changes and disrupted ecosystems</thought>
<thought>The loss of moonlight would affect nocturnal animals and human culture</thought>
<thought>Combining all these effects, we'd see a cascade of environmental changes</thought>
</thoughts>

The disappearance of the moon would have far-reaching consequences for Earth...
[detailed answer follows]

System Messages

The model continues to support various system prompts:

Basic prompt:

{"role": "system", "content": "You are a helpful assistant. Think before responding."}

Specific thought count:

{"role": "system", "content": "You are a helpful assistant. Think 3 thoughts before responding."}

Standard helper:

{"role": "system", "content": "You are a helpful assistant."}

Technical Details

Base Architecture

Base Model: Llama 3.2 3B
Initial Training: 2,500 carefully selected examples with up to 6 levels of thought chains
Thought Selection: Multi-level thought generation with external ranking system

RL Fine-tuning

Framework: OpenRLHF
Algorithm: REINFORCE
Judge Model: Gemini 1.5 Flash 8B
Training Parameters:
- Actor Learning Rate: 5e-7
- Critic Learning Rate: 9e-6
- Initial KL Coefficient: 0.01
- Batch Size: 128
- Max Epochs: 1
- Prompt/Generation Max Length: 1024
- BF16 Precision
- Flash Attention enabled
- Gradient Checkpointing
Training Data: OpenRLHF/prompt-collection-v0.1
Infrastructure: Ray distributed training with VLLM acceleration

What's It Good For?

The model excels at tasks requiring careful thinking and high-quality outputs:

✅ Breaking down complex problems with logical progression ✅ Step-by-step mathematical solutions with clear explanations ✅ Detailed analysis with well-structured arguments ✅ Clear and appropriate explanations of complicated concepts ✅ Well-reasoned decision-making with supporting evidence

Limitations

May still occasionally overthink simple problems
Bounded by base Llama 3.2 3B model capabilities
Not suitable for critical decisions without human oversight
Could generate irrelevant thought chains in edge cases
RL training might lead to occasional reward hacking behaviors

Example Usage

from transformers import AutoModelForCausalLM, AutoTokenizer

model = AutoModelForCausalLM.from_pretrained("ericflo/Llama-3.2-3B-COT-v3.0")
tokenizer = AutoTokenizer.from_pretrained("ericflo/Llama-3.2-3B-COT-v3.0")

messages = [
    {"role": "system", "content": "You are a helpful assistant. Think 3 thoughts before responding."},
    {"role": "user", "content": "How would you teach a child to ride a bike?"}
]

input_ids = tokenizer.apply_chat_template(messages, return_tensors="pt")
output = model.generate(input_ids, temperature=1.0)
response = tokenizer.decode(output[0])

Citation

@misc{thought-ranked-llama-v3,
  title={Thought-Ranked Llama 3.2 v3: RL-Optimized Hierarchical Chain-of-Thought Generation},
  author={[Eric Florenzano]},
  year={2024},
  howpublished={\url{https://huggingface.co/ericflo/Llama-3.2-3B-COT-v3}}
}

Acknowledgments

This model builds on the Llama 3.2 3B base model from Meta and incorporates RL training using Google's Gemini 1.5 Flash 8B as a judge. Special thanks to the open-source AI community for their contributions to chain-of-thought prompting techniques and reinforcement learning frameworks.

Downloads last month: 8

Safetensors

Model size

3B params

Tensor type

BF16

Model tree for ericflo/Llama-3.2-3B-COTv3

Base model

meta-llama/Llama-3.2-3B-Instruct

Quantized

(455)

this model

Quantizations

1 model

ericflo
/

Llama-3.2-3B-COTv3