A Chinese company has launched a CUDA-compatible GPU with a RISC-V CPU and a whopping 112GB HBM RAM – I bet Nvidia lawyers won’t be happy about that news
By Wayne Williams
Copyright techradar
Skip to main content
Tech Radar Pro
Tech Radar Gaming
Close main menu
the business technology experts
België (Nederlands)
Deutschland
North America
US (English)
Australasia
New Zealand
View Profile
Search TechRadar
Expert Insights
Website builders
Web hosting
Best web hosting
Best office chairs
Best website builder
Best antivirus
Expert Insights
Don’t miss these
A Chinese GPU vendor you’ve never heard of wants to challenge Nvidia and AMD in the pro market with 24GB VRAM – surely it cannot succeed?
End of Nvidia’s global dominance? Chipmaker summoned by Chinese government over security fears in H20 chips
Brave or foolhardy? Huawei takes the fight to Nvidia CUDA by making its Ascend AI GPU software open source
Watch how Chinese are converting thousands of Geforce RTX 5090 cards into AI accelerators and there’s nothing Nvidia can do to stop it
‘Could this be the RTX 6090’: Tech analyst reckons Nvidia’s just-announced AI GPU hides a tantalizing secret that could throw down a massive gauntlet to AMD
AMD drops hint that it’s planning a big GPU surprise for Nvidia soon, though I’m not fully convinced it’s the rumored RX 9080 XT
Nvidia has launched a GPU with 128GB of GDDR7 RAM but yeah, there’s no way it will sell one to us to run games
Ryzen AI Max+ 395 CPU could be AMD’s sleeper hit against Nvidia’s AI dominance, as nearly 30 Strix Halo mini AI workstation models hit the market – including some rather funky ones
Here’s how Nvidia and AMD hardware are being used in surprising ways to build Nvidia’s fastest GPU ever
Chaos at DeepSeek as R2 launch crashes into hardware problems – rivals gain huge advantage
Intel could take a huge leap ahead of AMD in gaming GPUs, with multi-frame generation potentially on the horizon
Nvidia CEO Jensen Huang says talks with Trump to allow chips into China will take time
Chinese vendor launches first workstation PC with Intel’s fastest CPU and up to two Arc Pro B60 GPUs, possibly with 48GB of RAM each
AMD’s next-gen flagship GPU could rival Nvidia’s mighty RTX 5090 – and I couldn’t be more excited
Another dual Intel GPU card with 48GB VRAM launched as Arc Pro emerges as cheap Nvidia and AMD alternative for AI crowd
A Chinese company has launched a CUDA-compatible GPU with a RISC-V CPU and a whopping 112GB HBM RAM – I bet Nvidia lawyers won’t be happy about that news
Wayne Williams
27 September 2025
Innosilicon Technology’s Fenghua No.3 was built from the ground up
When you purchase through links on our site, we may earn an affiliate commission. Here’s how it works.
(Image credit: Innosilicon Technology)
China’s Innosilicon launches Fenghua No.3 GPU with RISC-V CPU integration
Fenghua No.3 claims CUDA compatibility, 112GB HBM memory, and ray tracing
New GPU targets AI workloads, gaming, CAD, medical imaging, and 8K displays
Innosilicon Technology has introduced a new GPU in China which combines a RISC-V CPU with a graphics processor claiming compatibility with Nvidia’s CUDA platform.
The Fenghua No.3 (or Fantasy III as the English translation on the chip shows), the company’s latest flagship, comes equipped with over 112GB of HBM memory and targets a wide range of computing workloads, from AI to gaming.
Unlike its predecessors, which were based on PowerVR IP, the Fenghua No.3 draws on RISC-V architecture and is described by the company as a fully home-grown design.
You may like
A Chinese GPU vendor you’ve never heard of wants to challenge Nvidia and AMD in the pro market with 24GB VRAM – surely it cannot succeed?
End of Nvidia’s global dominance? Chipmaker summoned by Chinese government over security fears in H20 chips
Brave or foolhardy? Huawei takes the fight to Nvidia CUDA by making its Ascend AI GPU software open source
CUDA compatibility
At launch, representatives avoided giving detailed specifications, but stressed the chip’s support for multiple applications, including scientific computing, CAD, and commercial entertainment.
Fenghua No.3 was also billed as the first GPU capable of native DICOM support for medical imaging without specialized displays.
CUDA compatibility is easily the most striking claim, as the technology is proprietary to Nvidia, and few rival GPUs have attempted to replicate compatibility.
If the claims are accurate, Fenghua No.3 could prove attractive to researchers and developers already relying on Nvidia’s ecosystem, but I can’t imagine Nvidia not taking some form of action to protect its platform and intellectual property.
Are you a pro? Subscribe to our newsletter
Sign up to the TechRadar Pro newsletter to get all the top news, opinion, features and guidance your business needs to succeed!
Contact me with news and offers from other Future brandsReceive email from us on behalf of our trusted partners or sponsorsBy submitting your information you agree to the Terms & Conditions and Privacy Policy and are aged 16 or over.
In demonstrations, the card was shown running games such as Tomb Raider and Valorant, but no performance benchmarks, resolution details, or frame rates were provided.
For professionals, the card supports modern APIs including DirectX 12, Vulkan 1.2, and OpenGL 4.6, as well as hardware-based ray tracing. It’s compatible with Windows, Android, UOS, Kylin, and other operating systems.
The GPU’s memory configuration is well suited to AI applications and a single card is said to handle language models of up to 72B parameters, while systems with eight cards reportedly scale to 685B.
You may like
A Chinese GPU vendor you’ve never heard of wants to challenge Nvidia and AMD in the pro market with 24GB VRAM – surely it cannot succeed?
End of Nvidia’s global dominance? Chipmaker summoned by Chinese government over security fears in H20 chips
Brave or foolhardy? Huawei takes the fight to Nvidia CUDA by making its Ascend AI GPU software open source
The company also claims compatibility with several Chinese-developed AI models, pointing to ambitions in the country’s domestic AI sector.
Innosilicon claimed support for 8K resolution across six monitors, along with the YUV444 color format, a feature aimed at design and video professionals.
China has been working toward self-sufficiency in semiconductor technology, and products like the Fenghua No.3 is another example of this, although the card’s real-world performance remains to be seen.
You might also like
Huawei debuts AI system that’s faster than the market leader, the GB200 NVL72
Potentially tens of thousands of faulty Nvidia AI chips end up in Chinese repair shops
Mysterious tech firm set to become one of the biggest PC vendors in China
Wayne Williams
Social Links Navigation
Wayne Williams is a freelancer writing news for TechRadar Pro. He has been writing about computers, technology, and the web for 30 years. In that time he wrote for most of the UK’s PC magazines, and launched, edited and published a number of them too.
You must confirm your public display name before commenting
Please logout and then login again, you will then be prompted to enter your display name.
A Chinese GPU vendor you’ve never heard of wants to challenge Nvidia and AMD in the pro market with 24GB VRAM – surely it cannot succeed?
End of Nvidia’s global dominance? Chipmaker summoned by Chinese government over security fears in H20 chips
Brave or foolhardy? Huawei takes the fight to Nvidia CUDA by making its Ascend AI GPU software open source
Watch how Chinese are converting thousands of Geforce RTX 5090 cards into AI accelerators and there’s nothing Nvidia can do to stop it
‘Could this be the RTX 6090’: Tech analyst reckons Nvidia’s just-announced AI GPU hides a tantalizing secret that could throw down a massive gauntlet to AMD
AMD drops hint that it’s planning a big GPU surprise for Nvidia soon, though I’m not fully convinced it’s the rumored RX 9080 XT
Latest in Pro
Watch out – hackers are using AI to make phishing emails even more convincing
Nearly 150,000 patient records exposed in major healthcare data breach – here’s what we know
This graph alone shows how global AI power consumption is getting out of hand very quickly – and it’s not just about hyperscalers or OpenAI
This devious malware has jumped from Meta over to Google Ads and YouTube to spread – here’s how to stay safe
Microsoft flags dangerous XCSSET macOS malware targeting developers – so be on your guard
I helped create Amazon’s Alexa – here’s why I think businesses should embrace humanity’s analogous imperfection to succeed
Latest in News
YouTube Music is testing a new feature that sounds a lot like Spotify’s AI DJ
New Samsung Galaxy Z Tri-fold leaks hint at key software and camera zoom features
You can now use Google’s viral Nano Banana AI directly in Photoshop
ICYMI: the week’s 7 biggest tech stories from Apple’s response to ‘Scratchgate’ and the ROG Xbox Ally’s price reveal
Is the DJI Mini 5 Pro really a legal nightmare? Here’s what the world’s big drone regulators told me
Quordle hints and answers for Sunday, September 28 (game #1343)
LATEST ARTICLES
The Nutribullet SmartSense is an excellent, powerful blender, though its ‘smart’ mode feels unnecessary
This graph alone shows how global AI power consumption is getting out of hand very quickly – and it’s not just about hyperscalers or OpenAI
YouTube Music is testing a new feature that sounds a lot like Spotify’s AI DJ
Best Buy launches a surprise 48-hour flash sale ahead of Prime Day – here are the 15 best deals I’d buy
The Sony A7 V could launch soon – here are 5 upgrades that could make it the perfect all-rounder
TechRadar is part of Future US Inc, an international media group and leading digital publisher. Visit our corporate site.
Contact Future’s experts
Terms and conditions
Privacy policy
Cookies policy
Advertise with us
Web notifications
Accessibility Statement
Future US, Inc. Full 7th Floor, 130 West 42nd Street,
Please login or signup to comment
Please wait…