The Strongest Evolution in 25 Years: Nvidia Demonstrates DLSS 5 Super-Resolution Technology, Jensen Huang Calls It the GPT Moment for Graphics

Tech Home, March 17 — At the 2026 GPU Technology Conference (GTC) held today (March 17), NVIDIA CEO Jensen Huang showcased DLSS 5 super-resolution technology in his keynote speech and announced that this technology will be available this fall.

Jensen Huang stated:

After 25 years of inventing programmable shaders at NVIDIA, we have once again redefined computer graphics. DLSS 5 is a GPT moment for graphics, blending handcrafted rendering with generative artificial intelligence to achieve a huge leap in visual realism while maintaining artists’ control over creative expression.

Huang said that DLSS 5 is the most disruptive breakthrough in computer graphics since real-time ray tracing technology was introduced in 2018. The core of DLSS 5 introduces a new real-time neural rendering model that can completely break down the barriers between traditional rendering and the physical real world, helping game developers achieve photo-realistic visuals that previously only Hollywood visual effects could reach.

Huang emphasized that DLSS 5 is not only a redefinition of computer graphics but also a “GPT moment” in the graphics field. The related images are attached below:

This technology cleverly combines traditional handcrafted rendering with cutting-edge generative AI. In practical operation, DLSS 5 extracts color and motion vectors from each game frame as input data, then uses AI models to inject realistic lighting and materials into the scene in real time. Additionally, this technology can operate in real time at up to 4K resolution, ensuring a smooth interactive experience for players.

While generative AI video models have previously been able to produce realistic images, they often only run offline, and the generated content lacks coherence and controllability. To address this issue, NVIDIA has conducted end-to-end deep training of DLSS 5’s AI model.

Now, analyzing a single game frame allows the AI to accurately understand complex scene semantics. Whether it’s the gloss of character hair, fabric textures, or subsurface scattering effects under the skin, the AI can reproduce them perfectly. It can also accurately handle complex lighting environments such as front light, backlight, or overcast conditions, maintaining high consistency across different frames.

To ensure creative freedom for artists, DLSS 5 also provides developers with fine control over parameters such as intensity, color grading, and masking. The technology has already received unprecedented support from the industry. Top global game developers including Bethesda, Capcom, Ubisoft, Tencent, and NetEase have announced integration.

Players will be able to experience this revolutionary visual feast in major upcoming titles such as “Starfield,” “Assassin’s Creed: Mirage,” and “Resident Evil: Requiem” this fall.