Instructions to use piskle/Qwen2.5-3B-Instruct-Polski-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 piskle/Qwen2.5-3B-Instruct-Polski-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 piskle/Qwen2.5-3B-Instruct-Polski-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf piskle/Qwen2.5-3B-Instruct-Polski-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 piskle/Qwen2.5-3B-Instruct-Polski-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf piskle/Qwen2.5-3B-Instruct-Polski-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 piskle/Qwen2.5-3B-Instruct-Polski-GGUF:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf piskle/Qwen2.5-3B-Instruct-Polski-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 piskle/Qwen2.5-3B-Instruct-Polski-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf piskle/Qwen2.5-3B-Instruct-Polski-GGUF:Q4_K_M

Use Docker

docker model run hf.co/piskle/Qwen2.5-3B-Instruct-Polski-GGUF:Q4_K_M

LM Studio
Jan
Ollama
How to use piskle/Qwen2.5-3B-Instruct-Polski-GGUF with Ollama:
```
ollama run hf.co/piskle/Qwen2.5-3B-Instruct-Polski-GGUF:Q4_K_M
```
Unsloth Desktop

How to use piskle/Qwen2.5-3B-Instruct-Polski-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf piskle/Qwen2.5-3B-Instruct-Polski-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": "piskle/Qwen2.5-3B-Instruct-Polski-GGUF:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use piskle/Qwen2.5-3B-Instruct-Polski-GGUF with Docker Model Runner:
```
docker model run hf.co/piskle/Qwen2.5-3B-Instruct-Polski-GGUF:Q4_K_M
```

Lemonade

How to use piskle/Qwen2.5-3B-Instruct-Polski-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull piskle/Qwen2.5-3B-Instruct-Polski-GGUF:Q4_K_M

Run and chat with the model

lemonade run user.Qwen2.5-3B-Instruct-Polski-GGUF-Q4_K_M

List all available models

lemonade list

Hermes Agent

How to use piskle/Qwen2.5-3B-Instruct-Polski-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 piskle/Qwen2.5-3B-Instruct-Polski-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 piskle/Qwen2.5-3B-Instruct-Polski-GGUF:Q4_K_M

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use piskle/Qwen2.5-3B-Instruct-Polski-GGUF with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf piskle/Qwen2.5-3B-Instruct-Polski-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 "piskle/Qwen2.5-3B-Instruct-Polski-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"

Model is unstable due to lacking training / Model niestabilny ze względu na niewystarczające szkolenie

A fine-tuned version of Qwen2.5 3B Instruct, made for (at least understanding) Polish. The model was trained using the PolQA dataset. Available in multiple quantization variants, ranging from Q2_K_L up to F16. It should be used only for basic translation and conversations (From Polish to English and vice versa). Training was done on relatively low-end hardware (M1 chip, 8GB of unified memory, and 256GB of internal storage) for a kind of short time (an hour and 10 minutes), hence the low quality. The model can use anywhere from 1.3 to 6.3GB of RAM (depending on quantization, here Q2_K_L versus F16 with a context window of 8192).

Dostrojony wariant Qwen2.5 3B Instruct, przeznaczony do (co najmniej zrozumienia) języka polskiego. Model został przetrenowany za pomocą zbioru danych PolQA. Dostępny w kilku wariantach kwantyzacji, od Q2_K_L, aż do F16. Model powinien być używany jedynie do podstawowego tłumaczenia i rozmów (z polskiego na angielski i zamiennie). Trening był wykonany na sprzęcie o całkiem niskiej mocy obliczeniowej (czip Apple M1, 8GB zunifikowanego RAMu, 256GB pamięci wewnętrznej) przez krótki czas (godzina i dziesięć minut), stąd niska jakość. Model może używać of 1.3 do 6.3GB RAMu (zależy od kwantyzacji, tutaj Q2_K_L porównany z F16 z okienkiem kontekstu 8192).

Downloads last month: 131

GGUF

Model size

3B params

Architecture

qwen2

Hardware compatibility

2-bit

3-bit

4-bit

5-bit

6-bit

8-bit

16-bit

Inference Providers NEW

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

Model tree for piskle/Qwen2.5-3B-Instruct-Polski-GGUF

Base model

Qwen/Qwen2.5-3B

Finetuned

Qwen/Qwen2.5-3B-Instruct

Quantized

(299)

this model

piskle
/

Qwen2.5-3B-Instruct-Polski-GGUF

Model tree for piskle/Qwen2.5-3B-Instruct-Polski-GGUF

Dataset used to train piskle/Qwen2.5-3B-Instruct-Polski-GGUF