NVIDIA Video Codec SDK
NVIDIA Video Codec SDK is a comprehensive set of APIs including high-performance tools, samples, and documentation for hardware-accelerated video encode and decode on Windows and Linux.
Hardware-Based Decoder and Encoder
NVIDIA GPUs contain one or more hardware-based decoders and encoders (separate from the CUDA cores) which provide fully accelerated hardware-based video decoding and encoding for several popular codecs. GPU hardware accelerator engines for video encoding (NVENC) and video decoding (NVDEC) support faster-than-real-time video processing, which makes them suitable for use in multiple applications, including video transcoding, video data compression/decompression for deep learning, game broadcasting, virtual desktops, cloud gaming, secure video playback, video decode for training AI models etc.
Encoder
NVIDIA GPUs hardware-accelerated video encoder (NVENC), provides video encoding for H.264, HEVC (H.265) and AV1 codecs.
NVIDIA Blackwell introduces 422 progressive and interlaced encode support in hardware, enabling professional use cases in media and entertainment, video editing and broadcast.
The software enhancements in Video Codec SDK v13.0 enable extending ultra-high quality (UHQ) mode to AV1 encoding. This makes AV1 encoding on NVIDIA Blackwell NVENC comparable to software AV1 encoding with ~3X throughput. In January 2026, Jetson Thor encode/decode support was added to the Video Codec SDK.
Click image to expand
Decoder
NVIDIA GPUs contain an on-chip hardware-accelerated video decoder (NVDEC), which provides video decoding for several popular codecs.
NVDEC supports hardware-accelerated decoding of the following video codecs on Windows and Linux platforms: MPEG-2, VC-1, H.264 (AVCHD), H.265 (HEVC), VP8, VP9, and AV1.
With NVIDIA Blackwell GPUs, NVDEC doubles the video decoding throughput for H.264 decoding and adds support for 422 decoding for H.264 and HEVC codecs. NVIDIA Blackwell introduces 422 progressive and interlaced decode support in hardware.
Click image to expand
Access NVIDIA GPU encode/decode matrix
How Video Codec SDK Works
Video Codec SDK is a comprehensive set of APIs, high-performance tools, sample applications, reusable code, and documentation for hardware-accelerated video encoding and decoding on Windows and Linux.
These C-style APIs are useful for high-performance encoding and decoding using NVENC and NVDEC, respectively. These APIs expose the hardware functionality, along with other commonly used codec features such as rate control, framerates, codec profile, format selection etc. Video Codec SDK also provides a set of reusable code in the form of C++ classes built on top of the NVENCODE/NVDECODE APIs, which applications can easily integrate.
The NVENC and NVDEC can also be accessed via low-level APIs such as DirectX and Vulkan Video and via python APIs using PyNvVideoCodec. For more details regarding DirectX, Vulkan Video and PyNvVideoCodec please click here.
Key Features
Encode, Decode Performance Multiplier
Hardware encode and decode video processing enables high stream density vs software codec per form factor. This reduces the $/stream and Watt/stream.
High-Speed Video Acceleration for AI Training and Inference
As AI models rely on ever larger video datasets, decode and data throughput can become limiting factors. NVDEC helps remove these bottlenecks and together with NVENC enhances data curation workflows for building high-quality video training datasets.
8K240 fps on Blackwell and SW Comparable Codec Quality
Up to 8K240 FPS achievable with 4 NVENCs on NVIDIA Blackwell. UHQ mode introduced in Blackwell for AV1 enables SW comparable quality with greater than SW fps.
Get Started With Video Codec SDK
To utilize the dedicated decoding and encoding hardware on your GPU within an existing application, integration can be achieved through FFmpeg. While FFmpeg is suitable for evaluation or rapid integration, it may not offer granular control over every encoder parameter. For low-level, granular control over various encode/decode parameters and direct access to the hardware decoder/encoder, the NVDECODE and NVENCODE APIs should be employed. This direct access is provided via the Video Codec SDK.
Header Files
The Header files contain definition and description of the interface constants, structure, enums and function prototypes used in the APIs.
Download SDK
For the full Video Codec 13.0 package, please review the license agreement and check the box below to indicate that you accept it and wish to proceed to the download.
Blackwell features support:
422, 422i, 420i 8b/10b H.264 encode, decode support
422 8b/10b HEVC encode, decode support
Multi-view HEVC (ME-HEVC) 420 8b/10b support
Ultra-high quality (UHQ) mode for AV1
Double H.264 decode throughput per NVDEC compared to previous generation
Decode memory optimizations
NVIDIA Jetson Thor support
More Resources
Ethical AI
NVIDIA believes Trustworthy AI is a shared responsibility and we have established policies and practices to enable development for a wide array of AI applications. When downloaded or used in accordance with our terms of service, developers should work with their supporting model team to ensure this model meets requirements for the relevant industry and use case and addresses unforeseen product misuse.
For more detailed information on ethical considerations for this model, please see the Model Card++ Explainability, Bias, Safety & Security, and Privacy Subcards. Please report security vulnerabilities or NVIDIA AI Concerns here.