Instructions to use RichardErkhov/tuanpasg_-_puffin-linear-coder-math-1-gguf with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use RichardErkhov/tuanpasg_-_puffin-linear-coder-math-1-gguf 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 RichardErkhov/tuanpasg_-_puffin-linear-coder-math-1-gguf:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf RichardErkhov/tuanpasg_-_puffin-linear-coder-math-1-gguf:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf RichardErkhov/tuanpasg_-_puffin-linear-coder-math-1-gguf:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf RichardErkhov/tuanpasg_-_puffin-linear-coder-math-1-gguf: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 RichardErkhov/tuanpasg_-_puffin-linear-coder-math-1-gguf:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf RichardErkhov/tuanpasg_-_puffin-linear-coder-math-1-gguf: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 RichardErkhov/tuanpasg_-_puffin-linear-coder-math-1-gguf:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf RichardErkhov/tuanpasg_-_puffin-linear-coder-math-1-gguf:Q4_K_M

Use Docker

docker model run hf.co/RichardErkhov/tuanpasg_-_puffin-linear-coder-math-1-gguf:Q4_K_M

LM Studio
Jan
Ollama
How to use RichardErkhov/tuanpasg_-_puffin-linear-coder-math-1-gguf with Ollama:
```
ollama run hf.co/RichardErkhov/tuanpasg_-_puffin-linear-coder-math-1-gguf:Q4_K_M
```
Unsloth Desktop

How to use RichardErkhov/tuanpasg_-_puffin-linear-coder-math-1-gguf with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf RichardErkhov/tuanpasg_-_puffin-linear-coder-math-1-gguf:Q4_K_M

Configure the model in Pi

# Install Pi:
npm install -g @earendil-works/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "RichardErkhov/tuanpasg_-_puffin-linear-coder-math-1-gguf:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use RichardErkhov/tuanpasg_-_puffin-linear-coder-math-1-gguf with Docker Model Runner:
```
docker model run hf.co/RichardErkhov/tuanpasg_-_puffin-linear-coder-math-1-gguf:Q4_K_M
```

Lemonade

How to use RichardErkhov/tuanpasg_-_puffin-linear-coder-math-1-gguf with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull RichardErkhov/tuanpasg_-_puffin-linear-coder-math-1-gguf:Q4_K_M

Run and chat with the model

lemonade run user.tuanpasg_-_puffin-linear-coder-math-1-gguf-Q4_K_M

List all available models

lemonade list

Hermes Agent

How to use RichardErkhov/tuanpasg_-_puffin-linear-coder-math-1-gguf with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf RichardErkhov/tuanpasg_-_puffin-linear-coder-math-1-gguf: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 RichardErkhov/tuanpasg_-_puffin-linear-coder-math-1-gguf:Q4_K_M

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use RichardErkhov/tuanpasg_-_puffin-linear-coder-math-1-gguf with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf RichardErkhov/tuanpasg_-_puffin-linear-coder-math-1-gguf:Q4_K_M

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 "RichardErkhov/tuanpasg_-_puffin-linear-coder-math-1-gguf:Q4_K_M" \
  --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"

YAML Metadata Warning:empty or missing yaml metadata in repo card

Check out the documentation for more information.

Quantization made by Richard Erkhov.

Github

Discord

Request more models

puffin-linear-coder-math-1 - GGUF

Model creator: https://huggingface.co/tuanpasg/
Original model: https://huggingface.co/tuanpasg/puffin-linear-coder-math-1/

Name	Quant method	Size
puffin-linear-coder-math-1.Q2_K.gguf	Q2_K	0.7GB
puffin-linear-coder-math-1.IQ3_XS.gguf	IQ3_XS	0.77GB
puffin-linear-coder-math-1.IQ3_S.gguf	IQ3_S	0.8GB
puffin-linear-coder-math-1.Q3_K_S.gguf	Q3_K_S	0.8GB
puffin-linear-coder-math-1.IQ3_M.gguf	IQ3_M	0.82GB
puffin-linear-coder-math-1.Q3_K.gguf	Q3_K	0.86GB
puffin-linear-coder-math-1.Q3_K_M.gguf	Q3_K_M	0.86GB
puffin-linear-coder-math-1.Q3_K_L.gguf	Q3_K_L	0.91GB
puffin-linear-coder-math-1.IQ4_XS.gguf	IQ4_XS	0.96GB
puffin-linear-coder-math-1.Q4_0.gguf	Q4_0	0.99GB
puffin-linear-coder-math-1.IQ4_NL.gguf	IQ4_NL	1.0GB
puffin-linear-coder-math-1.Q4_K_S.gguf	Q4_K_S	1.0GB
puffin-linear-coder-math-1.Q4_K.gguf	Q4_K	1.04GB
puffin-linear-coder-math-1.Q4_K_M.gguf	Q4_K_M	1.04GB
puffin-linear-coder-math-1.Q4_1.gguf	Q4_1	1.08GB
puffin-linear-coder-math-1.Q5_0.gguf	Q5_0	1.17GB
puffin-linear-coder-math-1.Q5_K_S.gguf	Q5_K_S	1.17GB
puffin-linear-coder-math-1.Q5_K.gguf	Q5_K	1.2GB
puffin-linear-coder-math-1.Q5_K_M.gguf	Q5_K_M	1.2GB
puffin-linear-coder-math-1.Q5_1.gguf	Q5_1	1.26GB
puffin-linear-coder-math-1.Q6_K.gguf	Q6_K	1.36GB
puffin-linear-coder-math-1.Q8_0.gguf	Q8_0	1.76GB

Original model description:

base_model:

Qwen/Qwen2.5-Coder-1.5B
Qwen/Qwen2.5-1.5B
Qwen/Qwen2.5-Math-1.5B library_name: transformers tags:
mergekit
merge

merge

This is a merge of pre-trained language models created using mergekit.

Merge Details

Merge Method

This model was merged using the TIES merge method using Qwen/Qwen2.5-1.5B as a base.

Models Merged

The following models were included in the merge:

Configuration

The following YAML configuration was used to produce this model:

models:
  - model: Qwen/Qwen2.5-Coder-1.5B
    parameters:
      density: 1
      weight: 0.9
  - model: Qwen/Qwen2.5-Math-1.5B
    parameters:
      density: 0.5
      weight: 0.1

merge_method: ties
base_model: Qwen/Qwen2.5-1.5B
parameters:
  normalize: true
  int8_mask: true
dtype: bfloat16

Downloads last month: 95

GGUF

Model size

2B params

Architecture

qwen2

Hardware compatibility

2-bit

3-bit

4-bit

5-bit

6-bit

8-bit

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Paper for RichardErkhov/tuanpasg_-_puffin-linear-coder-math-1-gguf

Resolving Interference When Merging Models

Paper • 2306.01708 • Published Jun 2, 2023 • 19