Adobe and NVIDIA have announced an expansion of their partnership to optimize Adobe’s flagship tools, including Adobe Photoshop, Premiere, Substance 3D, Frame.io and Brand Concierge. This is being done to deliver faster AI-powered creation, production, and personalization to customers.
Adobe and NVIDIA announced a strategic partnership back in March. That collaboration included GPU acceleration for media ingestion and intelligence in Frame.io, as well as ongoing work around Adobe Substance 3D, Adobe Firefly Foundry and 3D digital twin workflows for enterprise customers.
Adobe is now bringing the power of the NVIDIA RTX Spark superchip, which NVIDIA launched earlier this month, to its agentic creative applications Photoshop, Premiere and Substance 3D. RTX-accelerated and optimized to take full advantage of RTX Spark’s unified memory and AI compute, these applications are claimed to deliver up to 2x faster AI processing, editing, color, and effects.
Adobe has also rearchitected Photoshop around GPU-accelerated compositing to be faster and more responsive for creative tasks.
Some of the most demanding post-production workflows in Premiere, from playback and rendering to effects, color, and AI-powered tools, are RTX-accelerated and optimized for RTX Spark. This results in real, tangible speed gains for popular audio and video features in Premiere like Enhance Speech and Object Mask. For 3D artists using Substance 3D Painter, RTX Spark speeds up the heaviest steps around texturing.
Adobe is also using NVIDIA accelerated computing to speed cloud media processing and intelligent content discovery for Adobe Frame.io. The updated Frame.io pipeline uses NVIDIA video acceleration, including NVIDIA NVDEC for hardware-accelerated decoding and NVIDIA TensorRT with CUDA for content preprocessing and embedding generation.
According to Adobe, moving compute-intensive video decoding, preprocessing, and embedding generation onto NVIDIA GPUs lets Frame.io process media faster and make it searchable sooner, while supporting more responsive discovery across large content libraries.
Frame.io transcode and search workloads are claimed to provide approx. 5x faster embedding generation, a 4-5x improvement in P95 processing latency, and an approximately 80% smaller search proxy.
Adobe and NVIDIA also worked together to optimize Brand Concierge using NVIDIA Nemotron open models. This version will be available to enterprise customers later this year. NVIDIA has also selected Brand Concierge to drive personalized conversational experiences grounded in organizational data and content. This enables NVIDIA to tailor interactions on its website based on customer needs, while leveraging Nemotron to deliver fast, high-quality responses at scale.
