I used to work for Graphcore.
I used to work for Graphcore. Microsoft gave up on Graphcore fairly early on actually and I think it was a wise decision. There were a number of issues with Graphcore's GC1 and GC2 chips: * a bet on storing the entire model (and code) in 900MB of SRAM. Hell of a lot of SRAM but it only really works for small models and the world wants enormous models. * Blew it's weirdness budget by a lot. Everything is quite different so it's a significant effort to port software to it. Often you did get a decent speedup (like 2-10x) but I doubt many people thought that was worth the software pain. * The price was POA so normal people couldn't buy one. (And it would have been too expensive for individuals anyway.) So there was little grass roots community support and research. Nvidia gets that because it's very easy to buy a consumer 4090 or whatever and run AI on it. * Nvidia were killing it with Grace Hopper. GC3 was way more ambitious and supports a mountain of DRAM with crazy memory bandwidth so if they ever finish it maybe they'll make a comeback.
Posted by IshKebab on Oct 4, 2025, 7:09 AM
Tags
Related issues
- In the UK Amazon have now raised the price of general shipping to be 2/3 the price of a months prime subscription.
- Credit where credit is due, critics claiming "[Apple device] is too expensive for something that just does [basic function]" have been pr...
- I like to buy music.
- Meet K-Claw.
- Fable is way too expensive for basic software stuff.
- [Meta] Abnormal Codex usage/quota depletion and usage-accounting inconsistencies — cross-report tracker
Raise real-world problems, find people facing the same issue, and discuss solutions.2,000 issues0 comments
pricing (SaaS)
In the UK Amazon have now raised the price of general shipping to be 2/3 the price of a months prime subscription.
Credit where credit is due, critics claiming "[Apple device] is too expensive for something that just does [basic function]" have been pr...
I like to buy music.
Meet K-Claw.
Fable is way too expensive for basic software stuff.
[Meta] Abnormal Codex usage/quota depletion and usage-accounting inconsistencies — cross-report tracker
[FEATURE] Passkey (WebAuthn) sign-in for Claude accounts, across every surface
> How are Chinese companies financing their buildouts of AI infrastructure?
I used to work for Graphcore.
I used to work for Graphcore. Microsoft gave up on Graphcore fairly early on actually and I think it was a wise decision. There were a number of issues with Graphcore's GC1 and GC2 chips: * a bet on storing the entire model (and code) in 900MB of SRAM. Hell of a lot of SRAM but it only really works for small models and the world wants enormous models. * Blew it's weirdness budget by a lot. Everything is quite different so it's a significant effort to port software to it. Often you did get a decent speedup (like 2-10x) but I doubt many people thought that was worth the software pain. * The price was POA so normal people couldn't buy one. (And it would have been too expensive for individuals anyway.) So there was little grass roots community support and research. Nvidia gets that because it's very easy to buy a consumer 4090 or whatever and run AI on it. * Nvidia were killing it with Grace Hopper. GC3 was way more ambitious and supports a mountain of DRAM with crazy memory bandwidth so if they ever finish it maybe they'll make a comeback.
Comments (0)
No comments yet. Start the discussion.