Instructions to use DavidOKB/MathThink-Qwen-3.5-4B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use DavidOKB/MathThink-Qwen-3.5-4B with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="DavidOKB/MathThink-Qwen-3.5-4B",
	filename="MathThink-Qwen-3.5-4B_alpha16/MathThink-Qwen-3.5-4B_alpha16-Q8_0.gguf",
)

llm.create_chat_completion(
	messages = "No input example has been defined for this model task."
)

Notebooks
Google Colab
Kaggle
Local Apps

llama.cpp

How to use DavidOKB/MathThink-Qwen-3.5-4B with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf DavidOKB/MathThink-Qwen-3.5-4B:Q8_0
# Run inference directly in the terminal:
llama-cli -hf DavidOKB/MathThink-Qwen-3.5-4B:Q8_0

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf DavidOKB/MathThink-Qwen-3.5-4B:Q8_0
# Run inference directly in the terminal:
llama-cli -hf DavidOKB/MathThink-Qwen-3.5-4B:Q8_0

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 DavidOKB/MathThink-Qwen-3.5-4B:Q8_0
# Run inference directly in the terminal:
./llama-cli -hf DavidOKB/MathThink-Qwen-3.5-4B:Q8_0

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 DavidOKB/MathThink-Qwen-3.5-4B:Q8_0
# Run inference directly in the terminal:
./build/bin/llama-cli -hf DavidOKB/MathThink-Qwen-3.5-4B:Q8_0

Use Docker

docker model run hf.co/DavidOKB/MathThink-Qwen-3.5-4B:Q8_0

LM Studio
Jan
Ollama
How to use DavidOKB/MathThink-Qwen-3.5-4B with Ollama:
```
ollama run hf.co/DavidOKB/MathThink-Qwen-3.5-4B:Q8_0
```

Unsloth Studio new

How to use DavidOKB/MathThink-Qwen-3.5-4B 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 DavidOKB/MathThink-Qwen-3.5-4B 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 DavidOKB/MathThink-Qwen-3.5-4B to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for DavidOKB/MathThink-Qwen-3.5-4B to start chatting

Pi new

How to use DavidOKB/MathThink-Qwen-3.5-4B with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf DavidOKB/MathThink-Qwen-3.5-4B:Q8_0

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": "DavidOKB/MathThink-Qwen-3.5-4B:Q8_0"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use DavidOKB/MathThink-Qwen-3.5-4B with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf DavidOKB/MathThink-Qwen-3.5-4B:Q8_0

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 DavidOKB/MathThink-Qwen-3.5-4B:Q8_0

Run Hermes

hermes

Docker Model Runner
How to use DavidOKB/MathThink-Qwen-3.5-4B with Docker Model Runner:
```
docker model run hf.co/DavidOKB/MathThink-Qwen-3.5-4B:Q8_0
```

Lemonade

How to use DavidOKB/MathThink-Qwen-3.5-4B with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull DavidOKB/MathThink-Qwen-3.5-4B:Q8_0

Run and chat with the model

lemonade run user.MathThink-Qwen-3.5-4B-Q8_0

List all available models

lemonade list

MathThink-Qwen-3.5-4B / README.md

WYK

Update Readme.md and model card

19750fa verified about 1 month ago

preview code

raw

history blame contribute delete

1.54 kB

metadata

license: apache-2.0
tags:
  - math
  - reasoning
  - qwen
  - llama-cpp
  - gguf
  - lora
  - chain-of-thought
datasets:
  - nvidia/Nemotron-SFT-Math-v3
base_model:
  - Qwen/Qwen3.5-4B

Qwen3.5-4B Math Fine-Tuned (Nemotron-SFT-Math-v3)

This model is a fine-tuned version of Qwen3.5-4B, explicitly optimized for complex mathematical reasoning and Chain-of-Thought (CoT) problem solving. It was fine-tuned using the Nemotron-Math-v3 dataset with Parameter-Efficient Fine-Tuning (PEFT/LoRA).

Model Details

Base Model: Qwen/Qwen3.5-4B
Fine-Tuning Dataset: nvidia/Nemotron-SFT-Math-v3
Methodology: LoRA (Rank = 64, Alpha = 32 or Alpha = 16). The lora_alpha scaling is specifically tuned to prevent catastrophic forgetting, ensuring the model retains conversational abilities while significantly enhancing mathematical logic.
Quantization: Safetensor format (F16) and GGUF formats (Q8_0)

Recommended Generation Parameters

Because this model leverages extensive Chain-of-Thought reasoning to solve math problems, the following generation parameters are highly recommended for the best performance:

{
  "temperature": 1.0,
  "top_p": 0.95,
  "repetition_penalty": 1.1
}

Note: A repetition_penalty of 1.1 is crucial to prevent the base model from occasionally falling into infinite generation loops on extremely long context windows.

Use Cases

Resolving complex math word problems (GSM8K).
Higher-level mathematical reasoning (MATH, AIME).
Step-by-step logic tracking and proofs.