The NVIDIA CUDA Profiling Tools Interface (CUPTI) is a dynamic library that enables the creation of profiling and tracing tools that target CUDA applications. CUPTI provides a set of APIs targeted at ISVs creating profilers and other performance optimization tools:

  • the Activity API,
  • the Callback API,
  • the Event API,
  • the Metric API,
  • the Profiler API,
  • the PC Sampling API,
  • the SASS Metric API, and
  • the Checkpoint API

Using these CUPTI APIs, independent software developers can create profiling tools that provide low and deterministic profiling overhead on the target system, while giving insight into the CPU and GPU behavior of CUDA applications. Normally packaged with the CUDA Toolkit, NVIDIA occasionally uses this page to provide CUPTI improvements and bug fixes between toolkit releases.

There is currently no CUPTI update to the CUDA Toolkit 12.5 Update 1. You may obtain the latest version of CUPTI by Downloading the CUDA Toolkit 12.5.1

Download the CUDA Toolkit 12.5 Update 1 Now Download the CUDA Toolkit 12.5 Now
Revision History

Key Features

  • Trace CUDA API by registering callbacks for API calls of interest
    • Full support for entry and exit points in the CUDA C Runtime (CUDART) and CUDA Driver
  • GPU workload trace for the activities happening on the GPU, which includes kernel executions, memory operations (e.g., Host-to-Device memory copies) and memset operations.
  • CUDA Unified Memory trace for transfers from host to device, device to host, device to device and page faults on CPU and GPU etc.
  • Normalized timestamps for CPU and GPU trace
  • Profile hardware and software event counters, including:
    • Utilization metrics for various hardware units
    • Instruction count and throughput
    • Memory load/store events and throughput
    • Cache hits/misses
    • Branches and divergent branches
    • Many more
  • Enables automated bottleneck identification based on metrics such as instruction throughput, memory throughput, and more
  • Range profiling to enable metric collection over concurrent kernel launches within a range
  • Metrics attribution at the high-level source code and the executed assembly instructions.
  • Device-wide sampling of the program counter (PC). The PC Sampling gives the number of samples for each source and assembly line with various stall reasons.
See the CUPTI User Guide for a complete listing of hardware and software event counters available for performance analysis tools.

Updates in CUDA Toolkit 12.5 Update 1

    Resolved Issues
    • Fixed an issue due to which resource callback CUPTI_CBID_RESOURCE_MODULE_PROFILED might not be issued when no other activity is enabled. This issue was introduced in the CUDA 12.4 Update 1 release.
    • Fixed a crash for the graph level tracing for CUDA graphs.

Updates in CUDA Toolkit 12.5

    New Features
    • Added APIs cuptiActivityEnableDriverApi and cuptiActivityEnableRuntimeApi to limit the tracing of CUDA APIs that are of interest. This can help in reducing the CUDA API tracing overhead.
    • Added new field cigMode to communicate the CUDA in Graphics (CIG) mode associated with the CUDA context. The activity record CUpti_ActivityContext2 is deprecated and it is replaced by a new activity record CUpti_ActivityContext3. Enum CUpti_ContextCigMode describes the supported CIG modes.
    • Added new field numMultiprocessors in the context activity record to communicate the number of multiprocessors assigned to the green context.
    • Tracing is supported for MPS (Multi-Process Service) on Tegra platforms.
    • Obfuscated symbols are provided for Linux x86_64 platform, it helps us in speeding up the debug process for issues like crash, hang etc. See the NVIDIA CUDA Toolkit Symbol Server technical blog for information on how to use obfuscated symbols. Symbol server address is: PC Sampling API from the header cupti_pcsampling.h which is supported on Volta and later GPU architectures.
    • Removed support for the PowerPC (ppc64le) architecture.
    Resolved Issues
    • Fixed Unified memory profiling on Heterogeneous Memory Management (HMM) and Address Translation Service (ATS) systems.


    Supported platforms
    • Linux x86_64[1]
    • Windows x86_64[1]
    • Linux aarch64 SBSA[1]
    • DRIVE OS QNX aarch64[2]
    • DRIVE OS Linux aarch64[2]
    [1] available in the CUDA Desktop Toolkit only
    [2] available in the Embedded or Drive toolkits only
    Supported NVIDIA GPU architectures
    • Activity and Callback APIs
      • All architectures supported by CUDA Toolkit
    • Event and Metric APIs
      • Volta
      • Pascal
      • Maxwell
    • Profiling and PC Sampling APIs
      • Hopper: GH100
      • Ada: AD10x
      • Ampere: A100 with Multi-Instance GPU, GA10x
      • Turing
      • Volta: GV100, GV10B
    CUDA Toolkit
      Please use the following drivers
      • 555.85 (Windows)
      • 555.42.06 (Linux)
      provided with CUDA Toolkit 12.5 Update 1 production release or a more recent version.



To provide feedback, request additional features, or report issues, please use the Developer Forums.

Installation Overview

When installing CUDA Toolkit 12.5 Update 1 and specifying options, be sure to select CUDA > Development > Tools > CUPTI.