NVIDIA's AI Gaming Assistant Project G-Assist Updated to Support RTX Graphics Cards with 6GB VRAM

Deep News
08/19

At Gamescom on August 15, NVIDIA announced that its Project G-Assist—an on-device AI gaming assistant—has been expanded to support all RTX GPUs with 6GB or more VRAM, including laptop graphics cards. This reduction in system requirements is made possible by NVIDIA's development of "a new, significantly more efficient AI model" that operates faster while reducing VRAM usage by 40% while maintaining response accuracy.

According to NVIDIA's announcement, the updated AI assistant will be available on August 19 through a Game Ready driver update in the NVIDIA app, with users able to download the G-Assist update directly from the NVIDIA app homepage.

This update also introduces a partnership with mod.io, facilitating the discovery and provision of G-Assist plugins built by users. Anyone with knowledge of JSON and Python scripting can build plugins for G-Assist, with NVIDIA providing a dedicated plugin builder for download.

Additionally, NVIDIA plans to release another update in September 2025, featuring specific tools for laptops to support NVIDIA BatteryBoost and Battery OPS functionality.

Project G-Assist was first announced by NVIDIA at Computex 2024 in Taipei as a free AI assistant designed for gamers. It is a small language model (SLM) that runs on RTX graphics cards in users' PCs, capable of providing "real-time diagnostics and recommendations to alleviate system bottlenecks, improve energy efficiency, optimize game settings, overclock GPUs, and more."

Project G-Assist allows users to issue "natural language commands" through text or voice, currently supporting only English, though more language support is expected if the project gains sufficient popularity. As mentioned, users can build their own Project G-Assist plugins, which will now be available directly through the application thanks to the mod.io partnership.

When Project G-Assist was initially launched, NVIDIA set high system requirements, requiring RTX 30, 40, or 50 series desktop GPUs with at least 12GB VRAM, which excluded even the latest Blackwell RTX 5000 series cards like the 8GB RTX 5060 Ti and RTX 5060. It also did not support laptop GPUs and NVIDIA RTX PRO GPUs.

These limitations have now been lifted, expanding Project G-Assist's reach, which is undoubtedly good news for more RTX players and developers. However, since Project G-Assist's launch, the AI assistant has remained in version 0.1 experimental stage, with room for improvement in actual execution. Testing has shown that G-Assist not only drastically reduces frame rates (from triple digits to single digits) but can even cause complete system crashes.

免責聲明:投資有風險,本文並非投資建議,以上內容不應被視為任何金融產品的購買或出售要約、建議或邀請,作者或其他用戶的任何相關討論、評論或帖子也不應被視為此類內容。本文僅供一般參考,不考慮您的個人投資目標、財務狀況或需求。TTM對信息的準確性和完整性不承擔任何責任或保證,投資者應自行研究並在投資前尋求專業建議。

熱議股票

  1. 1
     
     
     
     
  2. 2
     
     
     
     
  3. 3
     
     
     
     
  4. 4
     
     
     
     
  5. 5
     
     
     
     
  6. 6
     
     
     
     
  7. 7
     
     
     
     
  8. 8
     
     
     
     
  9. 9
     
     
     
     
  10. 10