DEV Community

El Bruno for Microsoft Azure

Posted on • Originally published at elbruno.com on

4

#SemanticKernel and GPT-4o: Image analysis labs in C#

Hi!

Introduction to GPT-4o

GPT-4o, developed by OpenAI, represents a significant leap forward in AI technology. Dubbed “omni” for its all-encompassing capabilities, GPT-4o is a multimodal model that can process and generate text, audio, and images.

It’s designed to facilitate more natural human-computer interactions, responding to audio inputs in as little as 232 milliseconds. This model is not only faster but also 50% cheaper to use in the API, making it a cost-effective solution for developers and businesses alike.

Sample GitHub Repository

I decided to test some of the new features, and created this repository with some samples using Semantic Kernel and the new GPT-4o model.

https://github.com/elbruno/gpt4ol-sk-csharp/

Repository Content

The repo describes the basic steps to set up and utilize the GPT-4o model with .NET. The repository includes sample code, and links to further reading materials.

The initial demo shows how to use GPT-4o to analyze the following image:

rpi shell console showing the rapsberry pi information with the tool neofetch

With the following output:


The image appears to be a screenshot of a terminal window running on a Raspberry Pi device. The user has executed the `neofetch` command with `sudo`, and the terminal displayed system information. Additionally, the `ollama list` command was executed, showing a list of local models.

Here's the breakdown of the terminal output:

### System Information (Neofetch Output)
- **OS:** Debian GNU/Linux 12 (bookworm) aarch64
- **Host:** Raspberry Pi 5 Model B Rev 1.0
- **Kernel:** 6.6.20+rpt-rpi-2712
- **Uptime:** 3 mins
- **Packages:** 694 (dpkg)
- **Shell:** bash 5.2.15
- **CPU:** 4 cores @ 2.400GHz
- **Memory:** 640MiB / 8052MiB

### Models List (Ollama List Command Output)
- **llama3:latest**
- **ID:** 71a106a91016
- **Size:** 4.7 GB
- **Modified:** 4 days ago
- **phi3:latest**
- **ID:** a2c89ceaed85
- **Size:** 2.3 GB
- **Modified:** 3 days ago

Enter fullscreen mode Exit fullscreen mode

I’ll keep updating the samples with more scenarios.

Happy coding!

Greetings

El Bruno

More posts in my blog ElBruno.com.


Top comments (0)

Image of Docusign

🛠️ Bring your solution into Docusign. Reach over 1.6M customers.

Docusign is now extensible. Overcome challenges with disconnected products and inaccessible data by bringing your solutions into Docusign and publishing to 1.6M customers in the App Center.

Learn more

👋 Kindness is contagious

Please leave a ❤️ or a friendly comment on this post if you found it helpful!

Okay