The Tool Desk
Outbyte PC Repair FREERepair Windows errors before they cause bigger problemsFix Now →Outbyte Driver Updater FREEFix the driver behind crashes, sound loss and screen glitchesFind Drivers →iTechGuides is reader-supported. When you buy through links on our site, we may earn an affiliate commission. As an Amazon Associate I earn from qualifying purchases. Learn more
SoftBank did acquire Graphcore—but not when the original headline first described the possibility. A contemporaneous report said the companies were in advanced talks; Graphcore confirmed the acquisition on July 11, 2024. It became a wholly owned SoftBank subsidiary and said it would continue operating under the Graphcore name. The announcement did not disclose the purchase price.
Did SoftBank buy Graphcore?
Yes. The news unfolded in two stages. Before the deal was confirmed, CRN summarized a Bloomberg report based on anonymous sources that SoftBank was in advanced talks to acquire Graphcore. CRN cautioned at the time that no final agreement was imminent and the discussions could fall apart. That was a report of negotiations, not confirmation of a completed transaction. CRN’s report on the talks
On July 11, 2024, Graphcore announced that SoftBank Group had acquired it. Graphcore said it would become a wholly owned subsidiary while continuing to operate under the Graphcore name. SoftBank later confirmed the July 2024 acquisition in its 2025 annual report. Graphcore’s acquisition announcement · SoftBank’s 2025 annual report
How much did SoftBank pay?
The official acquisition announcement reviewed for this article does not state a purchase price. CRN reported that Graphcore had a $2.8 billion valuation in 2020, but that is a historical valuation—not the amount SoftBank paid. The available sources do not establish the transaction price.
#1 Best Overall
- NVIDIA Volta GV100 Architecture — 4,608 CUDA Cores, 640 1st-Gen Tensor Cores delivering 14 TFLOPS FP32 and 112 TFLOPS deep learning performance for AI training, inference, HPC, and scientific computing workloads
- 32GB HBM2 ECC Memory — 900 GB/s Bandwidth — High-bandwidth memory on a 4096-bit bus with ECC error correction provides the memory capacity and throughput required for the largest AI models, simulations, and datasets
- PCIe 3.0 x16 Interface — 250W TDP — Standard PCIe Gen3 connectivity with passive cooling designed for enterprise rack server deployment in HPE ProLiant, Dell PowerEdge, and Supermicro platforms with adequate chassis airflow
- NVLink — Scale to 96GB Unified Memory — Connect two V100 GPUs via NVLink at 300 GB/s bi-directional bandwidth to scale GPU memory from 32GB to 96GB for larger AI training and HPC workloads
- Multi-Precision Computing — Supports FP64 (7 TFLOPS), FP32 (14 TFLOPS), FP16 (112 TFLOPS) and INT8 precision modes for flexible deployment across training, inference, and scientific simulation workloads
What does Graphcore make?
Graphcore is a Bristol-based business that designs, develops, and commercializes chips for artificial intelligence and machine learning. SoftBank’s Q1 FY2024 earnings presentation lists Graphcore as established in 2016 and identifies Bristol, UK, as its headquarters. It named Nigel Toon as CEO at that time. SoftBank’s Q1 FY2024 earnings presentation
What is an IPU?
Graphcore calls its processor an Intelligence Processing Unit, or IPU. CRN described the architecture as designed for AI applications and “fine-grained parallelism.” That describes Graphcore’s approach; it does not establish that IPUs outperform GPUs across workloads. A meaningful accelerator comparison would need evidence for the particular task, along with software support, system availability, and total system cost. The cited sources provide no comparable benchmark or pricing data for those factors. CRN’s report on Graphcore’s business and cloud access
Rank #2
- High-Performance AI Processing: The MX3 is designed to handle the most demanding AI computer vision workloads, delivering exceptional performance and efficiency.
- Flexible Integration: The MX3 can be easily integrated into your existing systems via its M.2 M-key form factor and support for Linux operating systems.
- Energy Efficient: The MX3 is designed to provide high performance while minimizing power consumption.
- Comprehensive Software Development Kit (SDK): The MX3 is supported by a comprehensive SDK that simplifies development and deployment.
- Hardware compatability: The MX3 is compatible with the PCI-SIG M.2 M-key 2280 Specification. It can be used with the Raspberry Pi 5 with a M-key 2280 HAT.
What do Graphcore’s historical business figures show?
CRN, citing Graphcore’s public filings with the UK government, reported that Graphcore’s 2022 revenue fell 46 percent to $2.7 million and its headcount fell 21.7 percent to 494 employees. CRN attributed slower system sales to deteriorating macroeconomic conditions. These are historical figures for 2022, not current measures of the company’s finances or staffing. CRN’s report on Graphcore’s historical financials
Could customers access Graphcore systems through the cloud?
Historically, CRN reported that Graphcore pursued cloud access to its IPU systems through Gcore in Europe and Paperspace in North America beginning in 2022. CRN also said Graphcore had ended its channel partner program while continuing to work with channel partners. This does not confirm that either provider currently offers Graphcore systems, or that the hardware is available to individual buyers. Current products, availability, geographic coverage, and program terms are not established by the cited sources. CRN’s report on cloud access and partners
Rank #3
- ✅Powered by 26 Tera-Operations Per Second (TOPS) Hailo-8 AI Processor. 2.5W typical power consumption
- ✅Scalable, enabling simultaneous processing of multi-streams & multi-models
- ✅Enabling real-time, low latency and high-efficiency AI inferencing on the edge devices
- ✅Supports TensorFlow, TensorFlow Lite, ONNX, Keras, Pytorch frameworks
- ✅Supports Linux and Windows. Supports the temperature range of -40°C to 85°C
What Graphcore and SoftBank said about the deal
In Graphcore’s July 11, 2024 announcement, co-founder and CEO Nigel Toon said: “Demand for AI compute is vast and continues to grow.” Toon also called the acquisition “a tremendous endorsement of our team and their ability to build truly transformative AI technologies at scale.” Those are Toon’s statements in the company’s announcement, not an independent assessment of Graphcore’s technology or performance. Graphcore’s announcement
Quick Recap
Best Value
- Professional AI & Creator Workstation: AMD Radeon AI PRO R9700 GPU with 32GB GDDR6 is engineered for AI development, professional content creation, and compute-intensive workloads.
- Massive 32GB Memory Capacity: 32GB of GDDR6 memory on a 256-bit bus provides ample bandwidth for large AI models, 8K video editing, and complex 3D rendering.
- Advanced RDNA 4 with AI Accelerators: 64 Compute Units with 3rd Gen Ray Tracing and dedicated 2nd Gen AI Accelerators for groundbreaking AI performance and visual computing.
- Professional Blower Cooling: Efficient single blower design exhausts heat directly out of the chassis, ideal for multi-GPU workstation and server configurations.
- Enterprise-Grade Thermal Solution: Vapor chamber heatsink with industrial Honeywell PTM7950 thermal interface material ensures reliable cooling under sustained professional loads.
Rank #4
- 48GB AI graphics accelerator
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
