# Prafulla Dhariwal

Prafulla Dhariwal is an Indian computer scientist and researcher at OpenAI, known for his contributions to diffusion models and the development of DALL-E 2, a generative AI model for text-to-image synthesis.

Prafulla Dhariwal is a computer scientist and researcher at [OpenAI](https://www.wikiprompt.org/wiki/openai), recognized for his work on [generative models](https://www.wikiprompt.org/wiki/generative-ai), particularly [diffusion models](https://www.wikiprompt.org/wiki/diffusion-models) and the development of DALL-E 2. His research has advanced the field of [artificial-intelligence](https://www.wikiprompt.org/wiki/artificial-intelligence), focusing on improving the quality and controllability of machine-generated images.

Dhariwal's early work at OpenAI included contributions to [deep-learning](https://www.wikiprompt.org/wiki/deep-learning) techniques and [neural networks](https://www.wikiprompt.org/wiki/neural-network). He gained prominence for his role in creating DALL-E 2, a model that generates images from textual descriptions, which became a landmark in [machine-learning](https://www.wikiprompt.org/wiki/machine-learning) and [computer vision](https://www.wikiprompt.org/wiki/computer-vision). His research on diffusion models, often in collaboration with colleagues, demonstrated that these models could outperform generative adversarial networks in image synthesis, leading to widespread adoption in the AI community.

## Education and Early Career

Dhariwal studied at the Massachusetts Institute of Technology, where he was affiliated with [MIT CSAIL](https://www.wikiprompt.org/wiki/mit-csail). During his time there, he engaged in research on [machine-learning](https://www.wikiprompt.org/wiki/machine-learning) and [deep-learning](https://www.wikiprompt.org/wiki/deep-learning). He later joined OpenAI, where he has been a key researcher in the [generative-ai](https://www.wikiprompt.org/wiki/generative-ai) division.

## Contributions to Diffusion Models

Dhariwal's most significant contribution is his work on diffusion models, a class of probabilistic models that generate data by gradually denoising random noise. In a 2021 paper, he and his co-authors introduced improvements to diffusion models, achieving state-of-the-art results on image generation benchmarks such as CIFAR-10 and ImageNet. This work established diffusion models as a viable alternative to GANs, sparking a wave of research in the field.

## DALL-E 2 and Text-to-Image Generation

Dhariwal was a core contributor to DALL-E 2, released by OpenAI in 2022. The model uses a diffusion-based approach to generate high-resolution images from text prompts, with capabilities such as inpainting and outpainting. DALL-E 2's success highlighted the potential of [large language models](https://www.wikiprompt.org/wiki/large-language-model) combined with diffusion techniques, influencing subsequent models like Stable Diffusion and Midjourney.

## Impact and Recognition

Dhariwal's work has been widely cited and has influenced both academic research and industry applications. He has been invited to speak at conferences and has received recognition within the AI community. His contributions have helped shape the direction of [generative-ai](https://www.wikiprompt.org/wiki/generative-ai), particularly in the area of text-to-image synthesis.

## Selected Publications

- Dhariwal, P., & Nichol, A. (2021). Diffusion Models Beat GANs on Image Synthesis. *Advances in Neural Information Processing Systems*.
- Ramesh, A., Dhariwal, P., et al. (2022). Hierarchical Text-Conditional Image Generation with CLIP Latents. *arXiv preprint*.

## See Also

- [diffusion-models](https://www.wikiprompt.org/wiki/diffusion-models)
- dall-e-2
- [openai](https://www.wikiprompt.org/wiki/openai)
- [generative-ai](https://www.wikiprompt.org/wiki/generative-ai)

## References

- OpenAI research publications and technical reports.
- Conference proceedings from NeurIPS and ICML.

---
Source: https://www.wikiprompt.org/wiki/prafulla-dhariwal
License: CC BY-SA 4.0 (https://creativecommons.org/licenses/by-sa/4.0/)
Last updated: 2026-09-05T13:26:33.208145+00:00