Instructions to use thanhkt/Qwen2.5-1.5B-MathInstruct with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use thanhkt/Qwen2.5-1.5B-MathInstruct with Transformers:

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("thanhkt/Qwen2.5-1.5B-MathInstruct", dtype="auto")

llama-cpp-python

How to use thanhkt/Qwen2.5-1.5B-MathInstruct with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="thanhkt/Qwen2.5-1.5B-MathInstruct",
	filename="unsloth.F16.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 thanhkt/Qwen2.5-1.5B-MathInstruct with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf thanhkt/Qwen2.5-1.5B-MathInstruct:F16
# Run inference directly in the terminal:
llama-cli -hf thanhkt/Qwen2.5-1.5B-MathInstruct:F16

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf thanhkt/Qwen2.5-1.5B-MathInstruct:F16
# Run inference directly in the terminal:
llama-cli -hf thanhkt/Qwen2.5-1.5B-MathInstruct:F16

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 thanhkt/Qwen2.5-1.5B-MathInstruct:F16
# Run inference directly in the terminal:
./llama-cli -hf thanhkt/Qwen2.5-1.5B-MathInstruct:F16

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 thanhkt/Qwen2.5-1.5B-MathInstruct:F16
# Run inference directly in the terminal:
./build/bin/llama-cli -hf thanhkt/Qwen2.5-1.5B-MathInstruct:F16

Use Docker

docker model run hf.co/thanhkt/Qwen2.5-1.5B-MathInstruct:F16

LM Studio
Jan
Ollama
How to use thanhkt/Qwen2.5-1.5B-MathInstruct with Ollama:
```
ollama run hf.co/thanhkt/Qwen2.5-1.5B-MathInstruct:F16
```

Unsloth Studio new

How to use thanhkt/Qwen2.5-1.5B-MathInstruct 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 thanhkt/Qwen2.5-1.5B-MathInstruct 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 thanhkt/Qwen2.5-1.5B-MathInstruct to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for thanhkt/Qwen2.5-1.5B-MathInstruct to start chatting

Pi new

How to use thanhkt/Qwen2.5-1.5B-MathInstruct with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf thanhkt/Qwen2.5-1.5B-MathInstruct:F16

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": "thanhkt/Qwen2.5-1.5B-MathInstruct:F16"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use thanhkt/Qwen2.5-1.5B-MathInstruct with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf thanhkt/Qwen2.5-1.5B-MathInstruct:F16

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 thanhkt/Qwen2.5-1.5B-MathInstruct:F16

Run Hermes

hermes

Docker Model Runner
How to use thanhkt/Qwen2.5-1.5B-MathInstruct with Docker Model Runner:
```
docker model run hf.co/thanhkt/Qwen2.5-1.5B-MathInstruct:F16
```

Lemonade

How to use thanhkt/Qwen2.5-1.5B-MathInstruct with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull thanhkt/Qwen2.5-1.5B-MathInstruct:F16

Run and chat with the model

lemonade run user.Qwen2.5-1.5B-MathInstruct-F16

List all available models

lemonade list

Uploaded model

Developed by: thanhkt
License: apache-2.0
Finetuned from model : unsloth/Qwen2.5-Math-1.5B-Instruct-bnb-4bit

This qwen2 model was trained 2x faster with Unsloth and Huggingface's TRL library.

Dataset

The model was trained on the Nvidia-mathinstuct dataset, which consists of 100,000 rows. This dataset was specifically chosen to enhance the model's mathematical reasoning and instruction-following capabilities.

🤗 Hugging Face Transformers

Qwen2.5-Math can be deployed and infered in the same way as Qwen2.5. Here we show a code snippet to show you how to use the chat model with transformers:


from unsloth import FastLanguageModel
import torch
max_seq_length = 4096 # Choose any! We auto support RoPE Scaling internally!
dtype = None # None for auto detection. Float16 for Tesla T4, V100, Bfloat16 for Ampere+
load_in_4bit = True # Use 4bit quantization to reduce memory usage. Can be False.


model, tokenizer = FastLanguageModel.from_pretrained(
    model_name = "thanhkt/Qwen2.5-1.5B-MathInstruct", 
    max_seq_length = max_seq_length,
    dtype = dtype,
    load_in_4bit = load_in_4bit,
    # token = "hf_...", # use one if using gated models like meta-llama/Llama-2-7b-hf
)
alpaca_prompt = """Below...

### Instruct:
{}

### Input:
{}

### Output:
{}"""

FastLanguageModel.for_inference(model) # Enable native 2x faster inference
inputs = tokenizer(
[
    alpaca_prompt.format(
        """A company wants to make a pipeline from a point A on shore to a point B on an island. The island is 6km from the coast. The price to build an onshore pipeline is $50,000 per kilometer, and $130,000 per kilometer to build

underwater. B' is the point on the coast so that BB' is perpendicular to the coast. The distance from A to B' is 9km. Position C on section AB' so that when connecting pipes according to ACB, the amount is minimal. At that time, C is one paragraph away from A by:

A. 6.5km B. 6km C. 0km D.9km""", # instruction
        "", # input
        "", # output - leave this blank for generation!
    )
], return_tensors = "pt").to("cuda")

from transformers import TextStreamer
text_streamer = TextStreamer(tokenizer)
_ = model.generate(**inputs, streamer = text_streamer, max_new_tokens = 512)