Corespan Systems

We build photonic based computing systems using a disaggregated architecture and software control plane.

[Corespan Systems @CorespanSystems]

What if the computer you needed didn't have to fit inside the server you bought?

Mercury Rack composes pooled GPUs, xPUs, SSDs & FPGAs over native PCIe-over-optical into right-sized blocks: one 32-GPU domain, two 16s, or four 8s.

https://t.co/18fvgYjBiX

[Corespan Systems @CorespanSystems]

In the Lab: 8x AMD Radeon AI PRO R9700, One Fabric

AMD specs the R9700 for 4-way. We ran 8 as one fabric.

Every GPU-to-GPU pair: ~111 GB/s bidirectional, line-rate PCIe 5.0 x16. Flat topology, no cross-domain cliff.

ROCm + PyTorch. No NVLink, no CUDA. ~2× what NVIDIA targets for the RTX PRO 6000.

Full breakdown https://t.co/zycHJHFU7F

[Corespan Systems @CorespanSystems]

The asset a lender repossesses isn't a GPU. It's a rack. And most AI racks financed at scale are HGX boxes: 8 accelerators welded to one baseboard, one topology, one cooling design, one generation.
New from Bill Koss on why bankable compute runs through the rack, not just the runtime: https://t.co/4ypo0YVnKY

[Corespan Systems @CorespanSystems]

The most expensive GPU is the one you can't fully use.

Wrong ratio, and half the rack sits idle waiting for a job shaped like the box.

Scale-across composes the exact compute, memory & I/O each workload needs — then reclaims it.

Utilization is the multiplier.
https://t.co/EWq3b7Ta9N

[Corespan Systems @CorespanSystems]

The inference era won't be won by the biggest boxes. It'll be won by the fabric that composes GPUs, memory & I/O around each workload, then reclaims it for the next one.

Tokens per dollar, not GPUs per rack.
https://t.co/WxzDqM78cO

[Corespan Systems @CorespanSystems]

@grok Very cool.

[Corespan Systems @CorespanSystems]

We were excited to see this development...let the software teams cook!

[DeepSeek @deepseek_ai]

🚀 DeepSeek-V4-Flash Official API is now LIVE in public beta!

🔷 We’ve massively upgraded its Agent capabilities—benchmark scores are now far surpassing the V4-Pro-Preview. Check out the massive performance leap below! 👇
🔷 The official V4-Flash now natively supports the Responses API format and is fully adapted for Codex!

Check out the configuration details in our official API docs: https://t.co/smCwQZMeiq

[DeepSeek @deepseek_ai]

🚀 DeepSeek-V4-Flash Official API is now LIVE in public beta!

🔷 We’ve massively upgraded its Agent capabilities—benchmark scores are now far surpassing the V4-Pro-Preview. Check out the massive performance leap below! 👇
🔷 The official V4-Flash now natively supports the Responses API format and is fully adapted for Codex!

Check out the configuration details in our official API docs: https://t.co/smCwQZMeiq

[Corespan Systems @CorespanSystems]

Introducing Corespan Mercury Rack
https://t.co/oemrHB6FJW

[Corespan Systems @CorespanSystems]

Using the PRU 2500 with RTX 5090 GPUs and Kimi K3.

https://t.co/bXPXAm1tZY

@totheagi
@Michaelzsguo

[Corespan Systems @CorespanSystems]

From Trophy to Utility:
https://t.co/XBeJh2v8SJ

[Corespan Systems @CorespanSystems]

In the lab with Graid:
https://t.co/IVDGKSR41K

[Corespan Systems @CorespanSystems]

Does it make sense to use a gaming GPU?
https://t.co/2RAvXHJv8S

[Corespan Systems @CorespanSystems]

Before there was Kimi, there was this:
https://t.co/CCUgC2o2DU

[Corespan Systems @CorespanSystems]

Disaggregated SSDs beats putting drives in servers.
https://t.co/VHQwUyb8Ph

[Corespan Systems @CorespanSystems]

How photonic switching enters the data center.
https://t.co/U6K4LHQ6Cr

[Corespan Systems @CorespanSystems]

The Light is Arriving in your data center.
https://t.co/V4GUP1ilT6

[Corespan Systems @CorespanSystems]

Catching up on some prior blog posts. Here is when we introduced the PRU 2500 with RTX 5090 GPUs:
https://t.co/CnkoJ3Ol9d