AI   LLMs

DeepSeek vs ChatGPT 1 day study

Written by Ahmad Alobaid Updated 29 January 2025 2—minute read

Report my experience of using both DeepSeek and ChatGPT for explaining statistical concepts over the course of a day. I tested both models by asking for clarifications on various topics and engaging in discussions using identical prompts. While their responses were similar, I observed key differences in speed, conciseness, and feedback style. This post highlights those distinctions and explores which model might be better suited for different learning preferences.

Content

  1. Summary
  2. Objective & Setup
  3. Example Responses
    1. 1️⃣ Explaining a Concept
    2. 2️⃣ Asking a Follow-up Question
    3. 3️⃣ Handling an Invalid Formula/Result
  4. What I Found: Key Differences
    1. 🚀 Speed & Conciseness
    2. 🎯 Error Handling
    3. 📌 Overall Output
  5. Limitations
  6. Final Thoughts

Summary

I wanted to compare DeepSeek-V3 and ChatGPT (GPT-4o) to see how they perform in my use case: studying statistical concepts and using LLMs to explain them. While their responses were very similar, I did notice some key differences.

Objective & Setup

  • Objective: Compare DeepSeek and ChatGPT from a practical perspective
  • Use Case: Explain statistical concepts
  • Duration: One working day
  • Versions Tested:
    • DeepSeek: DeepSeek-V3
    • ChatGPT: GPT-4o

The approach was simple: I read through statistical content, and whenever I encountered a concept, I typed the exact same prompt into both models. I then asked follow-up questions and discussed the topics further to see how they responded.

I also tested both text and image-based prompts to explore how well they handled different inputs.

Example Responses

1️⃣ Explaining a Concept

DeepSeek's Response

ChatGPT's Response

💡 Observation: DeepSeek provides a more concise definition, whereas ChatGPT elaborates further.

2️⃣ Asking a Follow-up Question

DeepSeek's Response

ChatGPT's Response

💡 Observation: DeepSeek answers the question concisely, while ChatGPT offers additional context before addressing the query.

3️⃣ Handling an Invalid Formula/Result

DeepSeek's Response

ChatGPT's Response

💡 Observation: DeepSeek directly points out the mistake, while ChatGPT provides a more nuanced response with additional explanation.

What I Found: Key Differences

🚀 Speed & Conciseness

  • DeepSeek was slightly faster and tended to give shorter, more concise and direct answers.
  • ChatGPT took a more conversational approach, often elaborating more on explanations.

🎯 Error Handling

  • DeepSeek was more straightforward and direct when pointing out errors.
  • ChatGPT tended to disagree more gently, sometimes adding extra context before correcting a mistake.

📌 Overall Output

Both models produced similar content, but their style and level of detail differed.

  • If you prefer quick and to-the-point explanations, DeepSeek might be better.
  • If you like a more detailed, discussion-like response, ChatGPT feels more engaging.

Limitations

It’s important to note that your experience may vary depending on what you’re using these models for.

Final Thoughts

Both DeepSeek and ChatGPT are great tools for understanding statistical concepts, but they have different strengths:

✔️ Want a fast, concise response? Go with DeepSeek.
✔️ Prefer a more engaging, detailed explanation? ChatGPT is a better fit.

At the end of the day, the best model depends on your preferences and use case.

✍️ This content was written by the author and refined using both ChatGPT and DeepSeek.

Share this post!