[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-5840":3},{"id":4,"name":5,"fullName":6,"owner":7,"repo":5,"description":8,"homepage":9,"htmlUrl":10,"language":11,"languages":10,"totalLinesOfCode":10,"stars":12,"forks":13,"watchers":14,"openIssues":15,"contributorsCount":16,"subscribersCount":16,"size":16,"stars1d":17,"stars7d":18,"stars30d":19,"stars90d":16,"forks30d":16,"starsTrendScore":20,"compositeScore":21,"rankGlobal":10,"rankLanguage":10,"license":22,"archived":23,"fork":23,"defaultBranch":24,"hasWiki":25,"hasPages":23,"topics":26,"createdAt":10,"pushedAt":10,"updatedAt":33,"readmeContent":34,"aiSummary":35,"trendingCount":16,"starSnapshotCount":16,"syncStatus":36,"lastSyncTime":37,"discoverSource":38},5840,"LACT","ilya-zlobintsev\u002FLACT","ilya-zlobintsev","Linux GPU Configuration And Monitoring Tool","",null,"Rust",4931,126,28,68,0,8,37,267,30,102.81,"MIT License",false,"master",true,[27,28,29,30,31,32],"amdgpu","gpu","hactoberfest","linux","nvidia","rust","2026-06-12 04:00:27","# Linux GPU Control Application\n\u003Ca href=\"https:\u002F\u002Ftranslate.fedoraproject.org\u002Fengage\u002Flact\u002F\">\n\u003Cimg src=\"https:\u002F\u002Ftranslate.fedoraproject.org\u002Fwidget\u002Flact\u002Fsvg-badge.svg\" alt=\"Translation status\" \u002F>\n\u003C\u002Fa>\n\n\u003Cimg src=\"res\u002Fio.github.ilya_zlobintsev.LACT.png\" alt=\"icon\" width=\"100\"\u002F>\n\nThis application allows you to control your AMD, Nvidia or Intel GPU on a Linux\nsystem.\n\n| GPU info                          | Overclocking                      | Fan control                       |\n| ----------------------------------| ----------------------------------| ----------------------------------|\n| ![image](.\u002Fres\u002Fscreenshots\u002F1.png) | ![image](.\u002Fres\u002Fscreenshots\u002F2.png) | ![image](.\u002Fres\u002Fscreenshots\u002F3.png) |\n| Software info                     | Historical data                   |                                   |\n| ![image](.\u002Fres\u002Fscreenshots\u002F4.png) | ![image](.\u002Fres\u002Fscreenshots\u002F5.png) |                                   |\n\n### Features:\n\n- #### Detailed GPU information reporting\n  - Name and manufacturer\n  - VBIOS info\n  - VRAM info (Type\u002FManufacturer\u002FBus)\n  - Hardware unit info (CUs\u002FSMs\u002FEUs, ROP count)\n  - Resizable BAR status\n  - Vulkan features and extensions\n- #### Monitoring\n  - Configurable historical charts for power\u002Fthermals\u002Ffrequency\n  - Throttling info\n  - Data CSV export\n- #### Power configuration\n  - Power cap\n  - Power states (AMD only)\n- #### Thermals configuration\n  - Custom fan curves (AMD\u002FNvidia)\n  - GPU firmware thermal options such as thermal and acoustic target\u002Flimit (AMD RDNA3+ only)\n- #### Overclocking\n  - GPU\u002FVRAM clocks configuration\n  - GPU undervolting (via voltage offset on AMD, [indirectly](https:\u002F\u002Fgithub.com\u002Filya-zlobintsev\u002FLACT\u002Fwiki\u002FFrequently-asked-questions#how-to-undervolt-nvidia-gpus) on Nvidia)\n- #### Settings profiles\n  - Automatic profile activation based on running processes or gamemode status\n- #### [OpenTelemetry metrics exporter](.\u002Fdocs\u002FEXPORTER.md)\n\nGPU configuration is handled by a system service that does not depend on a graphical session (Wayland\u002FX11).\n\nThe service can also be used standalone with a config file, for example in headless scenarios.\n\n# Quick links\n\n- [Installation](#installation)\n- [Hardware support](https:\u002F\u002Fgithub.com\u002Filya-zlobintsev\u002FLACT\u002Fwiki\u002FHardware-Support)\n- [Frequently asked questions](https:\u002F\u002Fgithub.com\u002Filya-zlobintsev\u002FLACT\u002Fwiki\u002FFrequently-asked-questions)\n- [Enable overclocking on AMD](https:\u002F\u002Fgithub.com\u002Filya-zlobintsev\u002FLACT\u002Fwiki\u002FOverclocking-(AMD))\n- [Config file reference](.\u002Fdocs\u002FCONFIG.md)\n- [API](.\u002Fdocs\u002FAPI.md)\n- [Power profiles daemon note](#power-profiles-daemon-note)\n- [Recovery from a bad overclock](https:\u002F\u002Fgithub.com\u002Filya-zlobintsev\u002FLACT\u002Fwiki\u002FRecovering-from-a-bad-overclock)\n- [Metrics exporter](.\u002Fdocs\u002FEXPORTER.md)\n- [Contribute code](.\u002Fdocs\u002FCONTRIBUTING.md)\n- [Contribute translations](#localization)\n- [Support the project](#support-the-project)\n\n# Installation\n\n- Arch Linux: Install the package from official repositories: `pacman -S lact`\n  (or `lact-git` from AUR for development builds).\n- Debian\u002FUbuntu\u002FDerivatives: Download a .deb from\n  [releases](https:\u002F\u002Fgithub.com\u002Filya-zlobintsev\u002FLACT\u002Freleases\u002F).\n\n  It is only available on Debian 12+ and Ubuntu 22.04+ as older versions don't\n  ship gtk4.\n- Fedora: use the\n  [Copr repository](https:\u002F\u002Fcopr.fedorainfracloud.org\u002Fcoprs\u002Filyaz\u002FLACT\u002F), or\n  download an RPM from\n  [releases](https:\u002F\u002Fgithub.com\u002Filya-zlobintsev\u002FLACT\u002Freleases\u002F).\n- Bazzite\u002FFedora Atomic: Use the Flatpak.\n- Gentoo: Available in\n  [GURU](https:\u002F\u002Fgithub.com\u002Fgentoo\u002Fguru\u002Ftree\u002Fmaster\u002Fsys-apps\u002Flact).\n- OpenSUSE: an RPM is available in\n  [releases](https:\u002F\u002Fgithub.com\u002Filya-zlobintsev\u002FLACT\u002Freleases\u002F).\n\n  Only tumbleweed is supported as leap does not have the required dependencies\n  in the repos.\n- NixOS: There is a package available in\n  [nixpkgs](https:\u002F\u002Fsearch.nixos.org\u002Fpackages?channel=unstable&from=0&size=50&sort=relevance&type=packages&query=lact).\n- Solus: Available in the offical repository: `eopkg it lact`\n- Flatpak (universal): Available on [Flathub](https:\u002F\u002Fflathub.org\u002Fapps\u002Fio.github.ilya_zlobintsev.LACT) and in [releases](https:\u002F\u002Fgithub.com\u002Filya-zlobintsev\u002FLACT\u002Freleases\u002F).\n\n  See the [Flatpak documentation](.\u002Fflatpak\u002FREADME.md) for additional notes.\n- Docker (service only, no GUI): See [DOCKER.md](.\u002Fdocs\u002FDOCKER.md)\n- Build from source.\n\nNote: Nvidia support requires the Nvidia proprietary driver with CUDA libraries\ninstalled.\n\n## Development builds\n\nTo get latest fixes or features that have not yet been released in a stable\nversion, there are packages built from the latest commit that you can install\nfrom the\n[test release](https:\u002F\u002Fgithub.com\u002Filya-zlobintsev\u002FLACT\u002Freleases\u002Ftag\u002Ftest-build)\nor using the `lact-git` AUR package on Arch-based distros.\n\nNote: the date that GitHub shows next to the test release is not when the packages were built,\nthe actual date is specified next to the attached package files.\n\n# Usage\n\nEnable and start the service (otherwise you won't be able to change any\nsettings):\n\n```\nsudo systemctl enable --now lactd\n```\n\nYou can now use the GUI to change settings and view information.\n\n# Hardware support\n\nSee the\n[Wiki page](https:\u002F\u002Fgithub.com\u002Filya-zlobintsev\u002FLACT\u002Fwiki\u002FHardware-Support)\n\n# Configuration\n\nThere is a configuration file available in `\u002Fetc\u002Flact\u002Fconfig.yaml`. Most of the\nsettings are accessible through the GUI, but some of them may be useful to be\nedited manually (like `admin_group` and `admin_user` to specify who has access\nto the daemon)\n\nSee [CONFIG.md](.\u002Fdocs\u002FCONFIG.md) for more information.\n\n**Socket permissions setup:**\n\nBy default, LACT uses either ether the `wheel` or `sudo` group (whichever is\navailable) for the ownership of the unix socket that the GUI needs to connect\nto.\n\nOn most desktop configurations (such as the default setup on Arch-based, most\nDebian-based or Fedora systems) this includes the default user, so you do not\nneed to configure this.\n\nHowever, some systems may have different user configuration. In particular, this\nhas been reported to be a problem on OpenSUSE.\n\nTo fix socket permissions in such configurations, edit `\u002Fetc\u002Flact\u002Fconfig.yaml`\nand under the `daemon` section either:\n\n- Set `admin_user` to your username\n- Set `admin_group` to a group that your user is a part of Then restart the\n  service (`sudo systemctl restart lactd`).\n\n# Overclocking (AMD)\n\nSome functionality requires enabling an option in the amdgpu driver, see the\n[wiki page](https:\u002F\u002Fgithub.com\u002Filya-zlobintsev\u002FLACT\u002Fwiki\u002FOverclocking-(AMD)) for\nmore information.\n\n## Power profiles daemon note!\n\nIf you are using `power-profiles-daemon` (which is installed by default on many\ndistributions), by default it may override the amdgpu performance level setting\naccording to its own profile.\n\nWhen using LACT 0.7.5+ and power-profiles-daemon 0.30+, LACT will try to connect to power-profiles-daemon \nand automatically disable the conflicting amdgpu action in ppd to avoid this conflict.\n\nIf running older versions, you can resolve this manually by creating a file at\n`\u002Fetc\u002Fsystemd\u002Fsystem\u002Fpower-profiles-daemon.service.d\u002Foverride.conf` with the\nfollowing contents:\n\n```\n[Service]\nExecStart=\nExecStart=\u002Fusr\u002Flibexec\u002Fpower-profiles-daemon --block-action=amdgpu_dpm\n```\n\nNote: the `\u002Fusr\u002Flibexec` path might be different on your system, check it in\n`systemctl status power-profiles-daemon`\n\nSee https:\u002F\u002Fgithub.com\u002Filya-zlobintsev\u002FLACT\u002Fissues\u002F370 for more information.\n\n# Suspend\u002FResume\n\nAs some of the GPU settings may get reset when suspending the system, LACT will\nreload them on system resume. This may not work on distributions which don't use\nsystemd, as it relies on the `org.freedesktop.login2` DBus interface.\n\n# Building from source\n\nDependencies:\n\n- rust 1.93+\n- gtk 4.6+\n- libadwaita 1.5+\n- git\n- pkg-config\n- clang\n- make\n- hwdata\n- libdrm\n\nOptional Dependencies:\n- vulkan-tools\n- clinfo\n\nCommand to install all dependencies:\n\n- Fedora:\n  `sudo dnf install rust cargo make git clang gtk4-devel libadwaita-devel libdrm-devel vulkan-tools clinfo`\n- Arch:\n  `sudo pacman -S --needed base-devel git clang make rust gtk4 libadwaita hwdata vulkan-tools clinfo`\n\nSteps:\n\n- `git clone https:\u002F\u002Fgithub.com\u002Filya-zlobintsev\u002FLACT && cd LACT`\n- `make`\n- `sudo make install`\n\nIt's possible to change which features LACT gets built with. To do so, replace\nthe `make` command with the following variation:\n\nHeadless build with no GUI:\n\n```\nmake build-release-headless\n```\n\n# Remote management\n\nIt's possible to have the LACT daemon running on one machine, and then manage it\nremotely from another.\n\nThis is disabled by default, as the TCP connection **does not have any\nauthentication or encryption mechanism!** Make sure to only use it in trusted\nnetworks and\u002For set up appropriate firewall rules.\n\nTo enable it, edit `\u002Fetc\u002Flact\u002Fconfig.yaml` and add `tcp_listen_address` with\nyour desired address and in the `daemon` section.\n\nExample:\n\n```yaml\ndaemon:\n  tcp_listen_address: 0.0.0.0:12853\n  log_level: info\n  admin_group: wheel\n  disable_clocks_cleanup: false\n```\n\nAfter this restart the service (`sudo systemctl restart lactd`).\n\nTo connect to a remote instance with the GUI, run it with\n`lact gui --tcp-address 192.168.1.10:12853`.\n\n# CLI\n\nThere is also a cli available.\n\n- List system GPUs:\n\n  `lact cli list-gpus`\n\n  Example output:\n\n  ```\n  10DE:2704-1462:5110-0000:09:00.0 (AD103 [GeForce RTX 4080])\n  ```\n- Getting GPU information:\n\n  `lact cli info`\n\n  Example output:\n\n  ```\n  $ lact cli info\n  GPU 10DE:2704-1462:5110-0000:09:00.0:\n  =====================================\n  GPU Model: NVIDIA GeForce RTX 4080 (0x10DE:0x2704)\n  Card Manufacturer: Micro-Star International Co., Ltd. [MSI] (0x1462)\n  Card Model: Unknown (0x5110)\n  Driver Used: nvidia 570.124.04\n  VBIOS Version: 95.03.1E.00.60\n  VRAM Size: 16376 MiB\n  GPU Family: Ada\n  Cuda Cores: 9728\n  SM Count: 76\n  ROP Count: 112 (14 * 8)\n  VRAM Type: GDDR6x\n  VRAM Manufacturer: Micron\n  L2 Cache: 65536 KiB\n  Resizeable bar: Enabled\n  CPU Accessible VRAM: 16384\n  Link Speed: 8 GT\u002Fs PCIe gen 3 x8\n  ```\n\n- Profiles\n  `lact cli profile [COMMAND]`\n\n  - List profiles:\n\n    `lact cli profile list`\n\n    Example output:\n\n    ```\n    Default\n    Gaming\n    Performance\n    Balanced\n    ```\n\n  - Get current Profile:\n\n    `lact cli profile get` or `lact cli profile`\n\n    Example output:\n\n    ```\n    Gaming\n    ```\n\n  - Set Profile:\n\n    `lact cli profile set \"Performance\"`\n\n    Example output:\n\n    ```\n    Performance\n    ```\n\n    - Auto switch profiles\n      `lact cli profile auto-switch [COMMAND]`\n\n        - Get auto-switch state:\n\n          `lact cli profile auto-switch get` or `lact cli profile auto-switch`\n\n          Example output:\n\n          ```\n          enabled\n          ```\n\n        - Enable auto switch:\n\n          `lact cli profile auto-switch enable`\n\n          Example output:\n\n          ```\n          enabled\n          ```\n\n        - Disable auto switch:\n\n          `lact cli profile auto-switch disable`\n\n          Example output:\n\n          ```\n          disabled\n          ```\n\nThe functionality of the CLI is quite limited. If you want to integrate LACT\nwith some application\u002Fscript, you should use the [API](.\u002Fdocs\u002FAPI.md) instead.\n\n# Reporting issues\n\nWhen reporting issues, please include your system info and GPU model.\n\nIf you're having an issue with changing the GPU's configuration, it's highly\nrecommended to include a debug snapshot in the bug report. You can generate one\nusing the option in the dropdown menu:\n\n![image](https:\u002F\u002Fgithub.com\u002Filya-zlobintsev\u002FLACT\u002Fassets\u002F22796665\u002F36dda5e3-981b-47e7-914e-6e29f30616b4)\n\nThe snapshot is an archive which includes the SysFS that LACT uses to interact\nwith the GPU.\n\nIf there's a crash, run `lact gui` from the command line to get GUI logs, check\ndaemon logs in `journalctl -u lactd` for errors, and see `dmesg` for kernel logs\nthat might include information about driver and system issues.\n\n# Localization\n\nYou can contribute translations to LACT using [Weblate](https:\u002F\u002Ftranslate.fedoraproject.org\u002Fengage\u002Flact\u002F).\n\n# Support the project\n\nIf you wish to support the project, you can do so via Patreon:\nhttps:\u002F\u002Fwww.patreon.com\u002FIlyaZlobintsev\n\nOr using cryptocurrency:\n- BTC: `12FuTXZzd5peGb7QfoRkXaLnbJ1DNVW4pP`\n- ETH: `0x80875173316aa6317641bfbc50644e7ca74d6b6d`\n- XMR: `42E93NZXM7STBUsnMRGNyxKryFVgpHKNP6aza94C5hn17j2W7zUnFHe7ASQzB3KorYYnsaVzWUyHHVYfcTLQRtB63qkv5jE`\n\n# Other tools\n\nHere's a list of other useful tools for AMD GPUs on Linux:\n\n- [CoreCtrl](https:\u002F\u002Fgitlab.com\u002Fcorectrl\u002Fcorectrl) - direct alternative to LACT,\n  provides similar functionality in addition to CPU configuration with a Qt UI\n- [amdgpu_top](https:\u002F\u002Fgithub.com\u002FUmio-Yasuno\u002Famdgpu_top) - tool for detailed\n  real-time statistics on AMD GPUs\n- [Tuxclocker](https:\u002F\u002Fgithub.com\u002FLurkki14\u002Ftuxclocker) - Qt overclocking tool,\n  has support for AMD GPUs\n","LACT 是一个用于 Linux 系统上的 GPU 配置和监控工具。它支持 AMD、NVIDIA 和 Intel 的显卡，提供了详细的 GPU 信息报告、实时监控、超频以及风扇控制等功能。该工具使用 Rust 语言编写，具有高效率和低资源占用的特点。用户可以通过 LACT 查看 GPU 的详细硬件信息、配置电源状态、设置自定义风扇曲线等，并且支持将数据导出为 CSV 格式。此外，LACT 还具备基于运行进程或游戏模式自动激活配置文件的能力。适用于需要对 GPU 进行精细调整与性能优化的场景，如游戏、图形渲染及机器学习等高性能计算领域。",2,"2026-06-11 03:05:09","top_language"]