Tomorrow
← Chips & Compute

Graphcore IPU

SoftBank · #16 most active of 18 in Chips & Compute

Compare →

IPU accelerators now inside SoftBank.

Current version

Bow IPU

Entry cost

System sales

Changes / 30d

0

The short answers

What is the latest version of Graphcore IPU?
The current shipped version of Graphcore IPU is Bow IPU.
How much does Graphcore IPU cost?
Graphcore IPU has no published flat entry price. System sales
What AI model does Graphcore IPU use?
Graphcore IPU runs primarily on n/a.

Source: SoftBank (no public source page recorded) · Reusable under CC BY 4.0 — cite Tomorrow

Capabilities

  • Wafer-on-wafer
  • Poplar SDK
  • Sparse compute

Pricing

System sales

Entry price over time

Not enough pricing history yet — we start charting from the second observation.

Geek mode

Models under the hood

  • n/a

    Provider undisclosed

    disclosed

Context window

—

Public API

no

Multi-model routing

no

Stack signals

  • TSMC
  • Poplar

Shared model stack

Other tracked products running on the same foundation models — a quick read on how much of the catalog moves when one of these models changes.

green disclosed · amber inferred · grey unattributed. Aliases are folded into one model; provider concentration counts only models with an established vendor.

Reported scale

Reference data. Each figure is whatever the source actually said — weekly users, downloads, revenue run-rate — with its own definition and date. These are not comparable between tools and are never used to rank anything.

No public usage figure on record for this product.

Also in Chips & Compute

All Chips & Compute
Tenstorrent Blackhole

Tenstorrent

16

Open RISC-V AI hardware

Version
Blackhole
Cost
from $999/unit
Model
Tensix cores
Google TPU Ironwood

Google

12

Inference-first TPU pods

Version
Ironwood (7th generation)
Cost
from $5.4/unit
Model
Gemini
Cerebras WSE-3

Cerebras

9

Wafer-scale inference and training

Version
WSE-3
Cost
Free tier / developer tier pricing available (contact/partner integrations)
Model
Codex-Spark
NVIDIA Blackwell

NVIDIA

9

GB200/B200 rack-scale AI systems

Version
Blackwell
Cost
Pricing is not publicly listed (enterprise hardware architecture/system)
Model
Blackwell
AWS Trainium2

AWS

5

Cloud-native training silicon

Version
Trainium2
Cost
Pricing not published; available via AWS EC2 instance pricing.
Model
Neuron compiler
Intel Gaudi 3

Intel

5

Cost-focused training and inference accelerator.

Version
1.24.1
Cost
Pricing not published. Available via Intel Tiber AI Cloud or OEM platforms.
Model
Accelerator

Elsewhere on the same models

Products in other segments that name one of Graphcore IPU's models in their stack.

Fivetran

Fivetran

Managed pipelines feeding the AI data layer.

Version
2025
Cost
Usage-based (MAR)
Model
n/a
Vertical Aerospace

Vertical AerospaceEU

UK eVTOL programme in piloted flight testing.

Version
VX4
Cost
Pre-revenue, orders booked
Model
n/a

Change history

No changes recorded yet.

Subscribe to Graphcore IPU changes

Graphcore IPU compared

Straight head-to-head pages against the busiest products in Chips & Compute.

Graphcore IPU vs Tenstorrent BlackholeGraphcore IPU vs Google TPU IronwoodGraphcore IPU vs Cerebras WSE-3Graphcore IPU vs NVIDIA BlackwellGraphcore IPU vs AWS Trainium2Graphcore IPU vs Intel Gaudi 3

How to cite this page

Free to cite and reuse under CC BY 4.0. Permalink: https://tomorrow.aliensquad.ai/tools/graphcore

APA
Tomorrow. (2026). Graphcore IPU — version, pricing and model stack [Data set entry]. AlienSquad. Retrieved 2026-09-17, from https://tomorrow.aliensquad.ai/tools/graphcore
BibTeX
@misc{tomorrow-tools-graphcore,
  author       = {{Tomorrow}},
  title        = {Graphcore IPU — version, pricing and model stack},
  year         = {2026},
  publisher    = {AlienSquad},
  howpublished = {\url{https://tomorrow.aliensquad.ai/tools/graphcore}},
  note         = {Accessed: 2026-09-17}
}