OServices/ Omini-pyauto-1.5b:latest

47 2 weeks ago

Omini-pyauto-1.5b is a fine-tuned Python automation code generation model built on top of Qwen2.5-Coder-1.5B-Instruct. It specializes in generating production-ready automation scripts for three major frameworks: Playwright, SeleniumBase, and Selenium. 

ollama run OServices/Omini-pyauto-1.5b

Details

2 weeks ago

84cfc7530082 · 986MB ·

model
qwen2
·
1.54B
·
Q4_K_M
system
You are an expert Python automation engineer. You write clean, production-ready scripts using Playwr
params
{ "num_ctx": 2048, "repeat_penalty": 1.1, "stop": [ "<|im_end|>", "<|im_
template
<|im_start|>system {{ .System }}<|im_end|> <|im_start|>user {{ .Prompt }}<|im_end|> <|im_start|>assi

Readme

Omini-pyauto-1.5b 🤖

A fine-tuned Python automation code generation model built on Qwen2.5-Coder-1.5B-Instruct, specialized for generating production-ready scripts for Playwright, SeleniumBase, and Selenium.

🚀 Quick Start

`bash ollama run OServices/Omini-pyauto-1.5b \`

💬 Example Usage

`“python from ollama import chat

response = chat( model=‘OServices/Omini-pyauto-1.5b’, messages=[{ ‘role’: ‘user’, ‘content’: ‘Write a Playwright script that scrapes all product names from a webpage.’ }], ) print(response.message.content) `”

🎯 What It Does Best

Task Example Prompt
Web scraping “Write a Playwright script to scrape prices from an e-commerce page”
Cloudflare bypass “Create a SeleniumBase script that bypasses Cloudflare protection”
Form automation “Write a Selenium bot that fills and submits a login form”
Page monitoring “Build a script that monitors a product page for price changes”
UI testing “Write a SeleniumBase pytest test that verifies login works”
Infinite scroll “Selenium script that scrolls infinitely and collects all items”

⚙️ Model Details

Property Value
Base model Qwen2.5-Coder-1.5B-Instruct
Fine-tuning QLoRA (r=16, float16)
Training data 1,215 balanced instruction-code pairs
Frameworks Playwright · SeleniumBase · Selenium
Format GGUF Q4_K_M
Size ~900MB
Context window 2048 tokens
License Apache 2.0

📦 Run via Python

`“python from ollama import chat

Web scraping example

response = chat( model=‘OServices/Omini-pyauto-1.5b’, messages=[{ ‘role’: ‘user’, ‘content’: ‘Write a SeleniumBase script to scrape headlines from a news site.’ }], ) print(response.message.content) `”

🔧 Run via REST API

`bash curl http://localhost:11434/api/chat -d '{ "model": "OServices/Omini-pyauto-1.5b", "messages": [{ "role": "user", "content": "Write a Playwright script that takes a screenshot of a URL." }] }' \`

💡 Tips for Best Results

  • Be specific: “Write a Playwright script that…” not just “scrape this site”
  • Mention the framework you want in your prompt
  • Ask for specific features: “with error handling”, “save to CSV”, “headless mode”
  • For anti-bot sites always specify: “use SeleniumBase with uc=True”

🏗️ Built With

  • Trained on Google Colab T4 (free tier)
  • Fine-tuned using QLoRA + PEFT
  • Dataset built from real GitHub automation scripts + hand-crafted seed examples + template generation
  • Converted to GGUF using llama.cpp

Built by OServices — first version, more improvements coming soon!


---

## 📢 Share It

Now that it's live, share it like this:

🚀 Just published my first fine-tuned LLM on Ollama!

Omini-pyauto-1.5b — generates Playwright, SeleniumBase and Selenium automation scripts locally.

✅ 100% offline ✅ ~900MB ✅ Free to use

ollama run OServices/Omini-pyauto-1.5b

https://ollama.com/OServices/Omini-pyauto-1.5b