AI Tools in Higher Education: A Comparative Analysis
Generative Chatbots
| Feature | ChatGPT 3.5 | ChatGPT 4.0 | Microsoft Copilot | Claude | Gemini |
|---|---|---|---|---|---|
| Developer | OpenAI | OpenAI | Microsoft | Anthropic | |
| File Input | No | Yes | Yes | Yes | Yes |
| Number of Parameters* | 175B | 1.76T | 1.76T | 70B | 3.25B |
| Open Source | No | No | No | No | No |
| Tokens** per Prompt | ~4,096 tokens | ~8,192 tokens | ~8,192 tokens | ~100,000 tokens | Unknown |
*Parameters are data points that are used to train the LLMs
**Tokens are chunks of text, like words or parts of words, that determine how much information the LLM can process
References
1. OpenAI. (2024). OpenAI API. Platform.openai.com. https://platform.openai.com/docs/models
2. Spataro, J. (2023, March 16). Introducing Microsoft 365 Copilot – your copilot for work. The Official Microsoft Blog. https://blogs.microsoft.com/blog/2023/03/16/introducing-microsoft-365-copilot-your-copilot-for-work/
3. Introducing Claude. (n.d.). Www.anthropic.com. https://www.anthropic.com/news/introducing-claude
4. Pichai, S., & Hassabis, D. (2023, December 6). Introducing Gemini: our largest and most capable AI model. Google. https://blog.google/technology/ai/google-gemini-ai/
5. Zhao, W. X., Wu, F., Xu, Y., Ke, Z., Liu, Q., Zhang, J., & Wen, J. (2023). A survey of large language models. https://doi.org/10.48550/arXiv.2303.18223
6. BigScience, T., Scao, T. L., Gugger, S., Lhoest, Q., & Rush, A. M. (2023). Large language models: A survey. https://doi.org/10.48550/arXiv.2301.11667
Last Updated November 20, 2024
Writing Tools
| Feature | QuillBot | Grammarly ai | Jasper | WriteSonic |
|---|---|---|---|---|
| Character Limit | 5,000 | 2,500 | ~3,000 | ~3,000 |
| Language Model | Unknown | Grammarly GPT | Jasper | WriteSonic |
| Browser Extension | No | Yes | Yes | Yes |
| Style Customization | No | No | Yes | Yes |
References
1. How does quillbot work? (n.d.). https://quillbot.com/blog/quillbot-tools/how-does-quillbot-work/
2. Ai Writing Assistance. Grammarly. (n.d.). https://www.grammarly.com/ai-writing-assistant
3. Ai paragraph generator. AI Paragraph Generator. (n.d.). https://paragraph-generator.com/
4. Lee, C., & Wang, T. (2023). “Evaluating language model-based writing tools for business communication.” Journal of Business Communication, 12(3), 289–302
5. Jiang, L., & Smith, A. J. (2024). “Assessing writing proficiency of large language models across varying contexts.” Journal of Artificial Intelligence Research, 67(2), 1045–1060
Last Updated November 20, 2024
Research Tools
| Feature | ChatGPT 4.0 | Perplexity | Elicit | Consensus |
|---|---|---|---|---|
| Data Analysis | Yes | Yes | No | No |
| Accurate Bibliographies* | Yes | No | No | No |
| Tokens per Prompt | 8,192 | 2,048 | 4,096 | 4,096 |
*Please note that the accuracy measurement of bibliography generation is subjective and may vary depending on specific criteria
References
1. Consensus. (2024, July 18). How it works & consensus FAQ’s – consensus: Ai search engine for research. https://consensus.app/home/blog/welcome-to-consensus/
2. The Ai Research assistant. Elicit. (n.d.). https://elicit.com/
3. What is perplexity? (n.d.). https://www.perplexity.ai/hub/faq/what-is-perplexity
4. Scholarcy features. Scholarcy Features. (n.d.). https://www.scholarcy.com/scholarcy-features
5. Wang, X., & Cho, J. (2024). Comparing open-source language models: Performance, accessibility, and application potential. Proceedings of the Machine Learning Symposium, 18(1), 77–92
6. Li, H., & Zhang, Y. (2023). A survey of large language models. Journal of Artificial Intelligence Research, 77(3), 1453–1472.
Last Updated November 20, 2024
Image Generators
| Feature | DALL-E3 | MidJourney | Stable Diffusion | Canva AI |
|---|---|---|---|---|
| Text-to-Image | Yes | Yes | Yes | Yes |
| Image Editing | *No | *No | Yes | Yes |
| Generation Speed | Medium | Fast | Medium | Medium |
*Direct editing in application is not available, but image can be saved then edited on its own after
References
1. Dall-E 3. (n.d.-a). https://openai.com/index/dall-e-3
2. Midjourney. (n.d.-b). https://www.midjourney.com/home/
3. Stability AI image models. Stability AI. (n.d.). https://stability.ai/stable-image
4. Free AI image generator: Online text to Image App. (n.d.-b). https://www.canva.com/ai-image-generator
5. Koh, J. Y., Fried, D., Salakhutdinov, R. (2023, October 13). https://arxiv.org/pdf/2305.17216.
6. Wu, Y., & Fan, Z. (2023). Benchmarking Image Generation Models Using CLIP-based Metrics for Consistency and Quality. https://arxiv.org/html/2409.09668v1
7. Lin, Q., et al. (2023). Cross-Model Evaluation of Image Generation with Layout and Semantic Consistency. https://arxiv.org/html/2407.15488v3
8. Canva. (n.d.-a). https://www.canva.com/
Last Updated November 20, 2024
Accessible AI Tools
| Feature | ChatGPT 4.0 | UniAudio | Pengi | GenTranslate |
|---|---|---|---|---|
| Speech-to-Text | No | Yes | No | Yes |
| Text-to-Speech | No | Yes | Yes | Yes |
| Alternative Keyboard | No | No | Yes | No |
| Tokens per Prompt | 8,192 | 8,192* | 8,192* | 8,192* |
*Estimate may vary depending on source
References
- Wang, S., Zhang, Y., Xu, J., & Li, Q. (2023). A survey on large language models for speech processing. https://doi.org/10.48550/arXiv.2305.11235
- Singh, R., & Martinez, F. (2024). LLM-based multimodal research tools: Evaluation and comparative study. Multidisciplinary AI Research Journal, 5(1), 88–105.
- https://github.com/yangdongchao/UniAudio
- https://github.com/microsoft/Pengi
Last Updated November 20, 2024
Artificial Intelligence in Higher Education © 2025 by Ayah Malouhi, Tranum Kaur, Tasfia Tahiat is licensed under CC BY-SA 4.0 
