Instructions to use FreeAIn/Pwen-3.5-2B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use FreeAIn/Pwen-3.5-2B with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="FreeAIn/Pwen-3.5-2B",
	filename="Qwen3.5-2B.Q4_K_M.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 FreeAIn/Pwen-3.5-2B 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 FreeAIn/Pwen-3.5-2B:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf FreeAIn/Pwen-3.5-2B:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf FreeAIn/Pwen-3.5-2B:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf FreeAIn/Pwen-3.5-2B: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 FreeAIn/Pwen-3.5-2B:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf FreeAIn/Pwen-3.5-2B: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 FreeAIn/Pwen-3.5-2B:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf FreeAIn/Pwen-3.5-2B:Q4_K_M

Use Docker

docker model run hf.co/FreeAIn/Pwen-3.5-2B:Q4_K_M

LM Studio
Jan
Ollama
How to use FreeAIn/Pwen-3.5-2B with Ollama:
```
ollama run hf.co/FreeAIn/Pwen-3.5-2B:Q4_K_M
```

Unsloth Studio

How to use FreeAIn/Pwen-3.5-2B 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 FreeAIn/Pwen-3.5-2B 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 FreeAIn/Pwen-3.5-2B to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for FreeAIn/Pwen-3.5-2B to start chatting

How to use FreeAIn/Pwen-3.5-2B with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf FreeAIn/Pwen-3.5-2B:Q4_K_M

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": "FreeAIn/Pwen-3.5-2B:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use FreeAIn/Pwen-3.5-2B with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf FreeAIn/Pwen-3.5-2B: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 FreeAIn/Pwen-3.5-2B:Q4_K_M

Run Hermes

hermes

Atomic Chat new

OpenClaw new

How to use FreeAIn/Pwen-3.5-2B with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf FreeAIn/Pwen-3.5-2B: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 "FreeAIn/Pwen-3.5-2B: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"

Docker Model Runner
How to use FreeAIn/Pwen-3.5-2B with Docker Model Runner:
```
docker model run hf.co/FreeAIn/Pwen-3.5-2B:Q4_K_M
```

Lemonade

How to use FreeAIn/Pwen-3.5-2B with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull FreeAIn/Pwen-3.5-2B:Q4_K_M

Run and chat with the model

lemonade run user.Pwen-3.5-2B-Q4_K_M

List all available models

lemonade list

Pwen 3.5

Pwen 3.5 is a finetuned version of Qwen3.5-2B specialized for study notes, explanations, and concise technical summaries. Finetuned by Pavel Hanzel using Unsloth on Apple Silicon.

Model Details

Attribute	Value
Base Model	`Qwen/Qwen3.5-2B` by Alibaba Cloud
Model Type	Causal Language Model
Finetune Method	QLoRA via Unsloth
Language	English
License	CC BY-SA 4.0 + Qwen License
Author	Pavel Hanzel
Release Date	July 2026

Files & Sizes

File	Format	Size	Use Case
`Qwen3.5-2B.Q4_K_M.gguf`	GGUF Q4_K_M	~1.3 GB	Recommended for Ollama, LM Studio, llama.cpp

Intended Use

Good for:

Generating structured study notes with headers + bullets
Explaining code, math, and technical concepts concisely
Summarizing documents into revision format
Q&A in educational contexts

Not good for:

Roleplay or creative fiction
Uncensored/unfiltered outputs - inherits Qwen safety
Non-English languages - trained on English only
Factual accuracy on events after 2025

Training Data

Finetuned on 202 curated prompt-response pairs covering: 3. Domains: JavaScript, Python, algorithms, physics, general text

LoRA config: r=16, alpha=16, dropout=0.01, target modules: q_proj, k_proj, v_proj, o_proj. Trained 300 Steps.

Recommend not using thinking mode. It can break and stuck itself inside thinking loop!

Downloads last month: 44

GGUF

Model size

2B params

Architecture

qwen35

Hardware compatibility

4-bit

Inference Providers NEW

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