Open Architecture AI Computing.
Zero Vendor Lock-In.
Custom AI chip solutions powered by Tenstorrent technology for automotive, aerospace, and drone applications where flexibility and independence matter.
QBlocks offers AI chip solutions based on Tenstorrent's open-source hardware. Unlike costly proprietary GPUs, our systems provide direct hardware access without licensing fees or vendor restrictions, giving you full control over your AI infrastructure.
Primary Applications
Automotive
In-vehicle infotainment, autonomous driving AI, ADAS, real-time sensor processing
Drones & UAVs
Autonomous flight AI, computer vision navigation, swarm coordination, BVLOS
Aerospace
Flight control AI, avionics computing, satellite systems, autonomous aircraft
Chip Solutions Catalog
↓ Blackhole™ Series - High-Performance AI Computing
Blackhole™ p100a
Contact us for prices →
The Blackhole™ p100a features Tenstorrent’s high performance Blackhole™ Tensix Processor, with 120 Tensix Cores and 28 GB of GDDR6, operating at up to 300W in an active-cooled form factor geared for desktop workstations. p100a is supported by Tenstorrent’s fully open source software stack, granting developers full access to the metal.
Software Capabilities
CPU – x86_64 architecture*
Motherboard – PCI Express 5.0 x16 slot, dual-slot width**
Memory – 64 GB
Storage – 100 GB (> 2 TB recommended)
Power Connectors – 12V-2x6
Total Board Power – Up to 300W
Power Supply – ATX 3.1 Certified
Operating Temperature Range (Die) – 0C - 95C
Operating System – Ubuntu version 22.04 (Jammy Jellyfish)
Blackhole™ p150a
Contact us for prices →
The Blackhole™ p150a features Tenstorrent’s high performance Blackhole™ Tensix Processor, with 140 Tensix Cores and 32 GB of GDDR6, operating at up to 300W in an active-cooled form factor geared for desktop workstations.
Software Capabilities
CPU – x86_64 architecture*
Motherboard – PCI Express 5.0 x16 slot, dual-slot width**
Memory – 64 GB
Storage – 100 GB (> 2 TB recommended)
Power Connectors – 12V-2x6
Total Board Power – Up to 300W
Power Supply – ATX 3.1 Certified or better
Operating Temperature Range (Die) – 0C - 95C
Operating System – Ubuntu version 22.04 (Jammy Jellyfish)
↓ Wormhole™ Series - Cost-Effective AI Solutions
Wormhole™ n150d
Contact us for prices →
Wormhole™ n150d features Tenstorrent’s flexible, scalable Wormhole™ Tensix Processor operating at up to 160W, offering superior performance for cost compared to traditional GPUs and broad data precision format support. The processor can network into a multichip mesh for workstations and servers (such as Galaxy) and is supported by Tenstorrent’s fully open source software stack, granting developers full access to the metal. n150d includes a 2.5-slot active cooling solution.
Software Capabilities
CPU – x86_64 architecture
Motherboard – PCI Express 4.0 x16 slot, Triple-slot width
Memory – 64 GB
Storage – 100 GB (≥2TB recommended)
Power Connectors – 4+4-pin EPS12V
Total Board Power – Up to 160W
Operating Temperature Range (Die) – 0C - 75C
Operating System – Ubuntu version 22.04 (Jammy Jellyfish)**
Internet Connection – Required for driver and stack installation
Wormhole™ n300d
Contact us for prices →
Wormhole™ n300d features two of Tenstorrent’s flexible, scalable Wormhole™ Tensix Processors operating at up to 300W, offering superior performance for cost compared to traditional GPUs and broad data precision format support. The processor can network into a multichip mesh for workstations and servers (such as Galaxy) and is supported by Tenstorrent’s fully open source software stack, granting developers full access to the metal. n300d includes a 2.5-slot active cooling solution.
Software Capabilities
CPU – x86_64 architecture
Motherboard – PCI Express 4.0 x16 slot, Triple-slot width
Memory – 64 GB
Storage – 100 GB (≥2TB recommended)
Power Connectors – 4+4-pin EPS12V
Total Board Power – Up to 300W
Operating Temperature Range (Die) –0C - 75C
Operating System – Ubuntu version 22.04 (Jammy Jellyfish)
Internet Connection – Required for driver and stack installation
Technology Roadmap
QBlocks is committed to delivering cutting-edge AI chip solutions with continuous advancement
6nm Process Technology
Advanced 6nm manufacturing for enhanced performance and power efficiency across Blackhole series.
4nm Process Technology
Next-generation 4nm process node delivering significant improvements in computational density and thermal efficiency.
2nm Process Technology
Industry-leading 2nm manufacturing technology for maximum performance per watt and scalability.
Ready to Own Your AI Data Now?
Talk to our team about your infrastructure needs — from a single deploymentto a national sovereign AI program.
Contact Us
Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.
Frequently Asked Questions
How does this compare to NVIDIA or AMD GPUs?
Comparable performance with open architecture, no licensing costs, and complete hardware-levelaccess. Better choice for applications requiring deep optimization or avoiding vendor lock-in.
When should I choose Blackhole over Wormhole?
Blackhole p100a (120 cores, 28GB) or p150a (140 cores, 32GB) for high-performance: Level 4-5autonomous driving, drone swarms, aerospace AI. Wormhole n150d (160W single processor) or n300d(300W dual processor) for cost-effective applications: infotainment, ADAS, edge computing.
Can I start with Wormhole and upgrade to Blackhole later?
Yes. Both use the same software stack. Code developed on Wormhole n150d/n300d runs onBlackhole p100a/p150a without rewrites. The p150a also supports multi-card linking for memorypooling and infinite scalability.
How long does integration take?
Typical timeline is 3-6 months from requirements to production deployment.
What memory configurations are available?
Blackhole p100a offers 28GB GDDR6, p150a offers 32GB GDDR6. The p150a supports multi-cardlinking to pool memory across cards for larger models. Wormhole series configurations vary byapplication - contact us for specific requirements.
What are the costs compared to proprietary GPUs?
Typically 40-60% less when factoring in hardware costs, zero licensing fees, and no vendorlock-in. Contact us for detailed quotes based on your configuration.