Nvidia and SPIL plan 15x expansion of co-packaged optics capacity by mid-2027
Nvidia CEO Jensen Huang announced at GTC Berlin that the company plans to expand its co-packaged optics (CPO) manufacturing capacity by 15 times by mid-2027. This expansion is a joint effort with packaging partner Siliconware Precision Industries (SPIL) and is intended to address the growing bottleneck in AI data centers, where traditional copper interconnects are reaching physical limits in…
Key points
- Nvidia and SPIL will expand co-packaged optics capacity 15x by mid-2027 and 30x by end of 2028.
- Nvidia committed $1.1 billion to US semiconductor testing with KYEC, starting April 2027.
- Amkor plans to grow interposer wafer production capacity by over 300% by end of 2028.
The push is centered on Nvidia’s new Spectrum-6 CPO switches, which integrate optical engines directly with silicon to improve efficiency and signal integrity. These switches are designed for Nvidia’s next-generation Vera Rubin AI factories. Huang framed this as part of a massive infrastructure buildout, noting that demand for optical components could rise 15-fold as AI models and accelerator clusters grow larger.
Beyond the optical expansion, Huang outlined broader manufacturing investments. Nvidia has committed $1.1 billion to advance US semiconductor testing capacity with partner KYEC, with production starting in April 2027. Additionally, partner Amkor plans to increase its interposer wafer production capacity by over 300% by the end of 2028. These moves signal a heavy bet by Nvidia and its partners on sustained demand for high-speed AI infrastructure components.
The story so far
3 episodes →- Nvidia and SPIL plan 15x expansion of co-packaged optics capacity by mid-2027this story
Nvidia and SPIL plan 15x expansion of co-packaged optics capacity by mid-2027
cryptobriefing.com · 24 October 2026
Loading the full article…
This text was published by cryptobriefing.com and written by Diego Almada Lopez. It is reproduced here with attribution so you can read it in full; the rights remain with the publisher. Read it at the source ↗
The headline, key points and digest above were generated by Digest AI's editorial model from the linked sources. Automated summaries can contain errors: the sources are the record. Spotted a mistake? Tell us. Published by Martin K., who runs Digest AI and handles corrections.
More in Hardware & Compute
All →- Zagita tool doubles RTX 30/40 GPU compute, driving used prices up · 1 src
- Investors watch AMD, Google, and Amazon challenge NVIDIA in AI chips · 2 src
- Google and Voltus expand virtual power plant deal for 1,070 megawatts in PJM grid · 1 src
- Cisco and NVIDIA unveil USFP module for AI factory networks · 1 src
- Ciena CEO says AI demand may outpace supply for its optical networking chips · 1 src
Comments
via GitHub Discussions