Fastvideo CUDA Software: Demo and Freeware Downloads

Updated: August 31, 2026

You can find below links for our demo and freeware software for image processing applications. This software is prepared for non-commercial use only. It is free for personal and educational use (including non-profit organizations). NO warranty and responsibility is provided by the authors for the consequences of using it. That software is based on our Image & Video Processing SDK.

Installation

Unzip all files into working folder and run console application from command line with parameters specified in Readme.txt. It's necessary to have NVIDIA GPU (Pascal, Volta, Turing, Ampere, Ada) installed to run the software on GPU. For the latest NVIDIA drivers visit NVIDIA Drivers Downloads. To get better performance use images with high resolution. If you need Linux version, please send us your request via contact form.

Software and Hardware Requirements

OS Windows-10 and 11 (64-bit)
Demo downloads for OS Linux Ubuntu and L4T are available upon request
Optimized for NVIDIA GPUs with Pascal, Volta, Turing, Ampere, Ada architectures
Available for NVIDIA Jetson TX2, NX and AGX Xavier, NX and AGX Orin
Download GPU SDK manual

Fast JPEG Codec on CUDA

Implementation is 100% compliant with JPEG Baseline standard
JPEG encoding and decoding for grayscale 8-bit and color 24-bit images (Baseline part of JPEG Standard)
JPEG encoding and decoding for 12-bit grayscale and 36-bit color images (Extension to JPEG Standard)
Supported image formats: BMP, PPM, PGM, JPG
Data input: 8/24-bit or 12/36-bit images from HDD/RAID/SSD
Data output: final compressed/uncompressed images at HDD/RAID/SSD
JPEG image quality in the range from 1 to 100
Subsampling modes 4:4:4, 4:2:2, 4:2:0
Performance measurements option

Fast JPEG Codec on CUDA

Size: 52.2 MB
Publisher: fastcompression.com
OS: Windows-10 (64-bit)
License: Time-limited demo
Built with CUDA 12.6
NVIDIA GPU: Pascal, Volta, Turing, Ampere, Ada

Fast JPEG2000 Codec on GPU

JPEG2000 encoding and decoding for grayscale 8/16-bit and color 24/48-bit images
Supported image formats: BMP, PPM, PGM, J2K, JP2
Data input: 8/10/12/14/16-bit monochrome or color images from HDD/RAID/SSD
Data output: final compressed images at HDD/RAID/SSD
Lossy JPEG2000 compression and decompression with CDF 9/7 wavelet
Lossless JPEG2000 compression and decompression with CDF 5/3 wavelet
Subsampling modes 4:2:0, 4:2:2, 4:4:4
Number of DWT levels up to 12
Codeblock size: 16×16, 32×32, 64×64
JPEG2000 image quality in the range from 1 to 100 (float value)
Quality layers – one
Progression order – LRCP (L = layer, R = resolution, C = component, P = position)
Modes of operation – single, batch, multithreaded-batch
Performance measurements option

Fast JPEG2000 GPU Codec

Size: 41.2 MB
Publisher: fastcompression.com
OS: Windows-10 (64-bit)
License: Time-limited demo
Built with CUDA 12.6
NVIDIA GPU: Pascal, Volta, Turing, Ampere, Ada

Fast Debayer on CUDA

All Bayer mosaic patterns for input data supported (RGGB, BGGR, GBRG, GRBG)
Mosaic algorithm to prepare CFA input image from any 24/48-bit PPM image
Read 8/16-bit input CFA image data from HDD/SSD/RAID in PGM format
Output format: 24/48-bit PPM
GPU Debayer algorithms: HQLI, L7, DFPD, MG
Performance measurements option for each algorithm

Fast Debayer on CUDA

Size: 135 MB
Publisher: fastcompression.com
OS: Windows-10 (64-bit)
License: Time-limited demo
Built with CUDA 12.6
NVIDIA GPU: Pascal, Volta, Turing, Ampere, Ada

ALTMapper demo software for CPU

Input image formats: JPG, TIFF, PNG, PPM, PGM
Visual perception image enhancement
Fast and accurate adaptive local tone mapping engine
Controlled Contrast Preservation and Visual Quality Improvement:
Local contrast
Micro contrast
Image detail preservation
Local Gain control
Over Gain mode for low-illumination cases
Moderate fall-off compensation
Soft cloud shadows suppression
Dehazing for aerial images
360 panoramas support
Automatic histogram smoothing

ALTMapper software on CPU

Size: 38 MB
Publisher: fastcompression.com
OS: Windows-10/11 (64-bit)
License: Time-limited demo

FastVCR software for machine vision cameras

Camera initialization (camera model, resolution, fps, etc.)
Camera control: input bit depth, image resolution and ROI, FPS and exposure time, Gain, camera bandwidth test
Optional modes: binning, decimation, dual gain, dual adc, etc.
Camera statistics: real fps, dropped frames, etc.
Image acquisition
Data unpacking
Bad pixel removal
White Balance / AWB
Adaptive Exposure and Gain control in real time
High quality demosaicing with MG algorithm
Color Correction with matrix profile or DCP profile
Highlight recovery
Exposure correction (brightness control)
Denoising: wavelet-based, bilateral, NLM
Rotation (90/180/270 degrees) and flip/flop
Resize (downscale and upscale)
Sharpening
Gamma transform
JPEG/MJPEG compression and storage to SSD
H.264/H.265/AV1 GPU-based encoding
Streaming via RTSP, storage to SSD
Real-time output to monitor
CLI application (optional)
Multicamera support (optional)
Support of other cameras

FastVCR software for XIMEA cameras

Size: 163 MB (Windows)
Publisher: fastcompression.com
OS: Windows-10/11 (64-bit), Ubuntu 22.04, L4T
License #1: Time-limited demo
NVIDIA GPU: Jetson TX2, NX/AGX Xavier, Orin
NVIDIA GPU: Pascal, Volta, Turing, Ampere, Ada

GPU SDK for image processing on CUDA

Image ingestion from a camera, frame grabber, HDD/SSD/RAM or GPU memory (PGM, BMP, PPM, JPG, YCbCr, RAW, DNG, byte array, OpenGL texture or PBO, etc.)
Raw data unpacking on GPU
Raw data linearization
Dark frame subtraction (FPN)
Flat-Field Correction (Shading Correction)
White Balance
Bad pixel suppression
Exposure correction
Raw Bayer Curves (master and RGB)
Raw Denoising
Demosaicing with HQLI, L7, DFPD, MG algorithms
Denoising for luma and chroma with wavelets, NLM, bilateral filters
Color Correction with matrix profile
Color Transforms: RGB->YCbCr, YCbCr->RGB, RGB->HSV, HSV->RGB, etc.
Composite or per-color component 8/10/12/14/16-bit 1D LUTs for tone mapping in RGB or HSV (Hue->Hue, Hue->Saturation, etc.)
3D LUT for RGB and HSV
RGB Histograms
Crop
Rotate to 90/180/270 degrees and flip/flop
Resize (downscale or upscale)
Remap (rotation to arbitrary angle, affine transform, perspective transform, projection, arbitrary mapping)
Gamma transform (composite or per-color component)
Sharpening with Unsharp Mask filter (USM)
Real-time output via OpenGL
JPEG codec (8/12 bits per channel)
Raw Bayer codec for lossy compression or decompression (8/12 bits per channel)
H.264 / H.265 encoding and decoding on GPU
Trace
Multiplexor
Stream-per-thread support for better performance
NVML support: GPU utilization, temperature, fan speed, power management
NVIDIA Jetson TX2, NX and AGX Xavier, AGX Orin support
FFmpeg filters and codecs on GPU (Remap, J2K, SDI in/out, etc.)

SDK for GPU Image Processing

Size: 157 MB (Windows)
Publisher: fastcompression.com
OS: Windows-10/11 (64-bit), Linux, L4T
License: Demo or Trial, ver. 0.23.1.0
Two builds — CUDA 12.6 and CUDA 13.3
CUDA 12.6 — Pascal, Volta, Turing, Ampere, Ada; Jetson TX2, NX/AGX Xavier, AGX Orin
CUDA 13.3 — Turing, Ampere, Ada; Jetson Thor

Open source projects from fastcompression.com

PGM2DNG Converter
GPU Camera Sample software for Windows-10/11, Linux Ubuntu 22.04 and L4T

About the author

Fyodor Serzhenko, PhD, is the founder and CEO of Fastvideo. He earned his PhD at the Moscow Institute of Physics and Technology (MIPT) in 1993. Since 2009 he has led the development of Fastvideo’s GPU-accelerated image codecs and ISP modules, including the Fastvideo SDK. Connect on LinkedIn.

Why you can trust these results

Fastvideo has built GPU-accelerated image processing software since 2009. Our codecs comply with the standards they implement — JPEG (ITU-T T.81 / ISO IEC 10918) and JPEG2000 (ITU-T T.800 / ISO IEC 15444) — and our tools are open source on GitHub. Every performance figure we publish is reproducible: see our benchmark methodology, download the Fastvideo SDK benchmark report (PDF), and measure it on your own GPU.