[GUEST ACCESS MODE: Data is scrambled or limited to provide examples. Make requests using your API key to unlock full data. Check https://lunarcrush.ai/auth for authentication information.]  Mike [@MikeLongTerm](/creator/twitter/MikeLongTerm) on x 8444 followers Created: 2025-07-21 14:01:44 UTC BREAKING @AMD & Tensorstack #AI Imaging $AMD ’s new Amuse XXX AI image generation is noticeably better than 3.0, and there’s no subscription fee AMD, in partnership with Tensorstack, has just launched Amuse 3.1, the latest version of its generative AI software suite. We previously saw Amuse XXX release earlier this year, working best on the newly released (at the time) RX 9070 XT with up to 4.3x faster performance. AMD says the new XXX version promises to deliver a “significant increase in image quality” thanks to now using “the world’s first block FP16 SD XXX Medium model”. It is optimized for Ryzen AI XDNA X NPUs. As such, it’s important to disclose that Amuse XXX is only supported on AI PCs. AMD notes that the generative AI software requires a Ryzen AI XXX series (laptop) or Ryzen AI MAX+ series processor with a XX TOPS XDNA X NPU and at least 24GB of RAM. Amuse XXX has improved image quality and less memory usage The BF16 datatype also comes with some inherent upgrades such as reduced memory usage. This means you can enjoy high-quality, local AI image generation with laptops fitted with a lower amount of RAM. Stable Diffusion (SD) XXX Medium will now run on 24GB laptops while consuming just 9GB of memory. AMD also provided us with some comparisons, and you can see the obvious improvements below on the new SD3 Medium data type compared to the version SDXL Turbo model. AMD also likes to point out the fact that there’s no subscription fee. A contrast to popular AI tools such as ChatGPT or Midjourney. The company claims to offer “stock-quality images with no subscription fees,” noting it is free for personal use and small to medium-sized enterprises (SME) under $X in annual revenue. AMD also points out you don’t need an internet connection to generate AI images – since it is locally powered by the NPU. Obviously, you’ll need an internet connection to download the software in the first place. “Image generation using SD X Medium is regulated by the Stability AI Community License and is free for personal use and for SMEs under $X million in annual revenue. Licensing requirements may change at the sole discretion of the third party.” In addition to the local AI image generation, XXX also supports rapid iteration with “no re-shoots”. In other words, you can quickly iterate on your image to create something closer to what you envisioned without having to start from scratch. AMD says “you can configure and custom-tailor the stock image to your brand, allowing for rapid design iteration and the speedy creation of marketing assets”. The company also advertises that the 2-stage pipeline powered by the XDNA X NPU can deliver high-resolution images, enhancing the output 2MP (1024×1024) image up to 4MP (2048×2048) image with upscaling. This makes them more suitable for print media.  XXXXX engagements  **Related Topics** [#ai](/topic/#ai) [xt](/topic/xt) [rx](/topic/rx) [generative](/topic/generative) [subscription](/topic/subscription) [coins ai](/topic/coins-ai) [$amd](/topic/$amd) [advanced micro devices](/topic/advanced-micro-devices) [Post Link](https://x.com/MikeLongTerm/status/1947295934905258180)
[GUEST ACCESS MODE: Data is scrambled or limited to provide examples. Make requests using your API key to unlock full data. Check https://lunarcrush.ai/auth for authentication information.]
Mike @MikeLongTerm on x 8444 followers
Created: 2025-07-21 14:01:44 UTC
BREAKING @AMD & Tensorstack #AI Imaging
$AMD ’s new Amuse XXX AI image generation is noticeably better than 3.0, and there’s no subscription fee
AMD, in partnership with Tensorstack, has just launched Amuse 3.1, the latest version of its generative AI software suite. We previously saw Amuse XXX release earlier this year, working best on the newly released (at the time) RX 9070 XT with up to 4.3x faster performance. AMD says the new XXX version promises to deliver a “significant increase in image quality” thanks to now using “the world’s first block FP16 SD XXX Medium model”. It is optimized for Ryzen AI XDNA X NPUs.
As such, it’s important to disclose that Amuse XXX is only supported on AI PCs. AMD notes that the generative AI software requires a Ryzen AI XXX series (laptop) or Ryzen AI MAX+ series processor with a XX TOPS XDNA X NPU and at least 24GB of RAM.
Amuse XXX has improved image quality and less memory usage
The BF16 datatype also comes with some inherent upgrades such as reduced memory usage. This means you can enjoy high-quality, local AI image generation with laptops fitted with a lower amount of RAM. Stable Diffusion (SD) XXX Medium will now run on 24GB laptops while consuming just 9GB of memory. AMD also provided us with some comparisons, and you can see the obvious improvements below on the new SD3 Medium data type compared to the version SDXL Turbo model.
AMD also likes to point out the fact that there’s no subscription fee. A contrast to popular AI tools such as ChatGPT or Midjourney. The company claims to offer “stock-quality images with no subscription fees,” noting it is free for personal use and small to medium-sized enterprises (SME) under $X in annual revenue. AMD also points out you don’t need an internet connection to generate AI images – since it is locally powered by the NPU. Obviously, you’ll need an internet connection to download the software in the first place.
“Image generation using SD X Medium is regulated by the Stability AI Community License and is free for personal use and for SMEs under $X million in annual revenue. Licensing requirements may change at the sole discretion of the third party.”
In addition to the local AI image generation, XXX also supports rapid iteration with “no re-shoots”. In other words, you can quickly iterate on your image to create something closer to what you envisioned without having to start from scratch. AMD says “you can configure and custom-tailor the stock image to your brand, allowing for rapid design iteration and the speedy creation of marketing assets”.
The company also advertises that the 2-stage pipeline powered by the XDNA X NPU can deliver high-resolution images, enhancing the output 2MP (1024×1024) image up to 4MP (2048×2048) image with upscaling. This makes them more suitable for print media.
XXXXX engagements
Related Topics #ai xt rx generative subscription coins ai $amd advanced micro devices