IPU Processors
The IPU (Intelligence Processing Unit) is a massively parallel processor co-designed with Poplar SDK for AI workloads.
What it does
The IPU (Intelligence Processing Unit) is a massively parallel processor co-designed with Poplar SDK for AI workloads. The Colossus MK2 GC200 features 1472 cores, 900MB In-Processor-Memory, and 250 teraFLOPS of AI compute at FP16.
- Category
- Compute and accelerators
- Public API
- Not offered
Not yet disclosed: deployment options, platforms, languages and support channels.
Pricing
Pricing isn't published
Graphcore doesn't list prices publicly. Ask them for a quote.
Security and compliance
No security details published yet
Graphcore hasn't published certifications or security practices for IPU Processors that we could find. Ask their team for audit reports.
Reviews
No reviews yet
Reviews come from people who use IPU Processors at work. Share how your team uses it to help other buyers decide.
Alternatives
NDP200
by Syntiant
Always-on vision sensor and speech processor built using the Syntiant Core 2 architecture.
Pricing Accelerator
by Earnix
Module within Earnix Price-It providing real-time dashboards, simulations, KPI tracking, and committee-ready reports for faster pricing decisions.
Quantum-Accelerated AI Servers
by Sygaldry Technologies, Inc.
Quantum-accelerated AI servers combining multiple complementary qubit modalities within a fault-tolerant architecture to provide exponential compute speedup for AI training and...
Reserved GPU capacity
by Pacific
Dedicated single-tenant bare-metal GPU capacity for training and inference.
RTX PRO Server
by NVIDIA
Universal data center platform for enterprise and industrial AI, powered by NVIDIA Blackwell.
Digital Optimized Hardware
by tNavigator
Hardware consultation and tailored selection of laptops, workstations, clusters, and on-demand cloud HPC for running reservoir models with maximum efficiency.
Is this your product?
Claim it to keep details accurate and submit evidence for the next ranking review.
Claim IPU Processors