Qwen 2.5 Coder 32B: Is This Best Open Weight Model Better than GPT-4o?

We look at the new Qwen-2.5 Coder 32B in this video. Its the new best open weight model that rivals GPT-4o and 3.5 Sonnet. I will show you how to build applications with this new SOTA model.


LINKS:
https://huggingface.co/Qwen/Qwen2.5-C...
https://qwenlm.github.io/blog/qwen2.5...
https://huggingface.co/chat/
https://aider.chat/docs/leaderboards/

Prompts used in the video:

Prompt #01: HTML Webpage:
"
Write a HTML code for a website in a single file with the following features.
There is a button in the middle of the page saying "click me"
When the button is clicked, show a random joke from a list of jokes. Make sure the jokes are funny.
When the button click, also make sure to randomly change the background of the page.
Add a random animation with each click.
"

Prompt #02: Text to image webapp:

Create an HMTL based web-app that takes a input from the user in the form of a text and will generate an image based on the text description using an external api and display the image to the user. Needs to have the following features.
When the page loads, there needs to be only one title input box in the middle of the page with greyed out text "Imagine something". This will be used for capturing user input.
When the user submits the text, send it to replicate api (documentation provided below) to use flux model to generate an image and display it to the user.
Add two more buttons at the bottom on the generate image, one for downloading the image and the other for regeneration of the image with the same prompt.
Keep the design modern and minimalist.

Provide detailed step by step instructions on how to run the app. Here is the python API documentation for Flux models on replicate:
curl -s -X POST \
-H "Authorization: Bearer $REPLICATE_API_TOKEN" \
-H "Content-Type: application/json" \
-H "Prefer: wait" \
-d $'{
"input": {
"prompt": "black forest gateau cake spelling out the words \\"FLUX SCHNELL\\", tasty, food photography, dynamic shot",
"go_fast": true,
"megapixels": "1",
"num_outputs": 1,
"aspect_ratio": "1:1",
"output_format": "webp",
"output_quality": 80,
"num_inference_steps": 4
}
}' \
https://api.replicate.com/v1/models/b..."
"

💻 RAG Beyond Basics Course:
https://prompt-s-site.thinkific.com/c...

Let's Connect:
🦾 Discord:   / discord  
☕ Buy me a Coffee: https://ko-fi.com/promptengineering
|🔴 Patreon:   / promptengineering  
💼Consulting: https://calendly.com/engineerprompt/c...
📧 Business Contact: [email protected]
Become Member: http://tinyurl.com/y5h28s6h

💻 Pre-configured localGPT VM: https://bit.ly/localGPT (use Code: PromptEngineering for 50% off).

Signup for Newsletter, localgpt:
https://tally.so/r/3y9bb0


00:00 Introduction to the Quen 2.5 Coder Series
00:46 Overview of the Quen 2.5 Models
01:40 QuenCoder Performance on Leaderboards
03:40 Technical Details and Configuration
04:46 Building a Simple Web Application
07:31 Creating a Text-to-Image Web App
12:58 Testing and Comparison with GPT-4
14:45 Conclusion and Final Thoughts

All Interesting Videos:
Everything LangChain:    • LangChain  

Everything LLM:    • Large Language Models  

Everything Midjourney:    • MidJourney Tutorials  

AI Image Generation:    • AI Image Generation Tutorials