Instructions to use dennisonb/qwen25-tax-3b with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use dennisonb/qwen25-tax-3b with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="dennisonb/qwen25-tax-3b",
	filename="qwen25-tax-3b-v3-q8_0.gguf",
)

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

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use dennisonb/qwen25-tax-3b with llama.cpp:

Install (macOS, Linux)

curl -LsSf https://llama.app/install.sh | sh
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf dennisonb/qwen25-tax-3b:Q8_0
# Run inference directly in the terminal:
llama cli -hf dennisonb/qwen25-tax-3b:Q8_0

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf dennisonb/qwen25-tax-3b:Q8_0
# Run inference directly in the terminal:
llama cli -hf dennisonb/qwen25-tax-3b: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 dennisonb/qwen25-tax-3b:Q8_0
# Run inference directly in the terminal:
./llama-cli -hf dennisonb/qwen25-tax-3b: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 dennisonb/qwen25-tax-3b:Q8_0
# Run inference directly in the terminal:
./build/bin/llama-cli -hf dennisonb/qwen25-tax-3b:Q8_0

Use Docker

docker model run hf.co/dennisonb/qwen25-tax-3b:Q8_0

LM Studio
Jan
Ollama
How to use dennisonb/qwen25-tax-3b with Ollama:
```
ollama run hf.co/dennisonb/qwen25-tax-3b:Q8_0
```

Unsloth Studio

How to use dennisonb/qwen25-tax-3b 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 dennisonb/qwen25-tax-3b 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 dennisonb/qwen25-tax-3b to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for dennisonb/qwen25-tax-3b to start chatting

How to use dennisonb/qwen25-tax-3b with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf dennisonb/qwen25-tax-3b: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": "dennisonb/qwen25-tax-3b:Q8_0"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use dennisonb/qwen25-tax-3b with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf dennisonb/qwen25-tax-3b: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 dennisonb/qwen25-tax-3b:Q8_0

Run Hermes

hermes

Atomic Chat new

OpenClaw new

How to use dennisonb/qwen25-tax-3b with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf dennisonb/qwen25-tax-3b:Q8_0

Configure OpenClaw

# Install OpenClaw:
npm install -g openclaw@latest
# Register the local server and set it as the default model:
openclaw onboard --non-interactive --mode local \
  --auth-choice custom-api-key \
  --custom-base-url http://127.0.0.1:8080/v1 \
  --custom-model-id "dennisonb/qwen25-tax-3b:Q8_0" \
  --custom-provider-id llama-cpp \
  --custom-compatibility openai \
  --custom-text-input \
  --accept-risk \
  --skip-health

Run OpenClaw

openclaw agent --local --agent main --message "Hello from Hugging Face"

Docker Model Runner
How to use dennisonb/qwen25-tax-3b with Docker Model Runner:
```
docker model run hf.co/dennisonb/qwen25-tax-3b:Q8_0
```

Lemonade

How to use dennisonb/qwen25-tax-3b with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull dennisonb/qwen25-tax-3b:Q8_0

Run and chat with the model

lemonade run user.qwen25-tax-3b-Q8_0

List all available models

lemonade list

Qwen2.5 Tax 3B — IRS Tax Code Expert

A 3B parameter model fine-tuned on the Internal Revenue Code using a 3-stage RL pipeline:

SFT (Supervised Fine-Tuning) on 16,909 RAG-grounded Q&A pairs from all 2,113 IRC sections
DPO (Direct Preference Optimization) with 1,311 hard-negative pairs + on-policy error correction
GRPO (Group Relative Policy Optimization) with citation accuracy reward signal

Training Pipeline

Stage	Data	Iterations	Key Metric
SFT	16,909 grounded pairs	1,000	Val loss: 0.765
DPO	1,311 preference pairs	500	Loss: 0.005
GRPO	16,909 prompts	300	Avg reward: 0.978

Model Versions

v1: Initial SFT + DPO + GRPO (basic reward)
v2: Improved grounded data + hard-negative DPO (best factual accuracy)
v3: + On-policy DPO + citation accuracy reward (best citation specificity)

Usage with Ollama

# Download the GGUF
wget https://huggingface.co/dennisonb/qwen25-tax-3b/resolve/main/qwen25-tax-3b-v3-q8_0.gguf

# Create Ollama model
ollama create qwen25-tax-3b -f Modelfile

# Run
ollama run qwen25-tax-3b "What is the standard deduction for a single filer?"

Evaluation Results

Model	5-Q Score	GRPO Reward	Notes
v1	2.5/5	0.605	Frequent hallucinations
v2	4.5/5	0.828	Best factual accuracy
v3	3.5/5	0.978	Best citation specificity

Limitations

3B model cannot reliably memorize all IRC section numbers and dollar thresholds
May hallucinate specific amounts (e.g., Section 179 limits)
Best used with RAG (retrieval-augmented generation) for production
Not a substitute for professional tax advice

Training Data

All training data was generated using RAG from the actual IRC statutory text:

Source: 2,113 IRC sections parsed from the US Code
Generation: GPT-4o-mini with actual statute text in context
Validation: Cross-reference checking, citation accuracy validation
Cost: ~$9 total API cost for full dataset generation

Built With

MLX - Apple Silicon native ML framework
Qwen2.5-3B-Instruct - Base model
Ollama - Local model deployment
OpenAI Batch API - Training data generation

Downloads last month: 9

GGUF

Model size

3B params

Architecture

qwen2

Hardware compatibility

8-bit

Video Preview

Reinforcement Learning

Model tree for dennisonb/qwen25-tax-3b

Base model

Qwen/Qwen2.5-3B

Finetuned

Qwen/Qwen2.5-3B-Instruct

Quantized

(251)

this model