4 ms·

Vsora Jotunn-8 5nm European inference chip

161 points by rdg42 10mo ago

all2 10mo ago
288GB RAM on board, and RISC V processors to enable the option for offloading inference from the host machine entirely. It sounds nice, but how much is it?
- rq1 10mo ago
  The next generation will include another processor to offload the inference from the RISC V processors used to offload inference from the host machine.
  - ddalex 10mo ago
    The next next generation will include memory to offload memory from the on chip memory to the memory on memory (also known as SRAM cache)
pclmulqdq 10mo ago
It needs a "buy a card" link and a lot more architectural details. Tenstorrent is selling chips that are pretty weak, but will beat these guys if they don't get serious about sharing. Edit: It kind of looks like there's no silicon anywhere near production yet. Probably vaporware.
- layer8 10mo ago
  Tapeout apparently completed last month, dev boards in early 2026: https://www.eetimes.eu/vsora-tapes-out-ai-inference-chip-for-data-centers/ https://www.eetimes.eu/vsora-tapes-out-ai-inference-chip-for...
- embedding-shape 10mo ago
  Nice wave they've been able to ride if it's vaporware, considering they're been at it for five years. Any guesses to why no one else seemingly see the obvious you see?
  - pclmulqdq 10mo ago
    Look at the CGI graphics and indications in their published material that all they have is a simulation. A It's all there without disclosing an anticipated release date. Even their product pages and their news page don't seem to have indications of this. Also, the 3D graphic of their chip on a circuit board is missing some obvious support pieces, so it's clearly not from a CAD model. Lots of chip startups start as this kind of vaporware, but very few of them obfuscate their chip timelines and anticipated release dates this much. 5 years is a bit long to tapeout, but not unreasonable.
    - embedding-shape 10mo ago
      > Even their product pages and their news page don't seem to have indications of this. This seems indicative enough for me, give or take a quarter or two probably, from the latest news post on their website: > VSORA is now preparing for full-scale deployment, with development boards, reference designs, and servers expected in early 2026. https://vsora.com/vsora-announces-tape-out-of-game-changing-inference-chip-putting-europe-at-the-forefront-of-data-center-ai/ https://vsora.com/vsora-announces-tape-out-of-game-changing-... Seems they have partners as well, who describe working together with a Taiwanese company as well. You never know, guess they could have gotten others to fall for their illusions too, it's not unheard of. But considering how long time something like this takes to bring to market, that they have dev-boards ready is months rather than years at least gives me enough to wait until then to judge them too harshly.
      - lukan 10mo ago
        "that they have dev-boards ready is months rather than years at least gives me enough to wait until then to judge them too harshly." So far, they just talk about it.
  - simondotau 10mo ago
    The company itself doesn't have to be vaporware in order for the product, as described, to be vaporware. And even if it's not true vaporware, there's still no independent evidence of anything. The design could end up being a bust, whether because of unexpected hardware limitations, or an immature software stack, or underwhelming thermals, or sheer economics.
unit149 10mo ago
[dead]
N_Lens 10mo ago
An FP8 performance of 3200TFLOPS is impressive, could be used for training as well as inference. "Close to theory efficiency" is a bold statement. Most accelerators achieve 60-80% of theoretical peak; if they're genuinely hitting 90%+, that's impressive. Now let's see the price.
bangaladore 10mo ago
I love that the JS loads so slow on first load that it just says "The magic number: 0 /tflops"
- unwind 10mo ago
  It loaded fine for me, but that slash before the unit was a bit smelly. :| Just a tiny edit, but it's a rather core part of their message so they should probably notice and format it correctly before publishing.
  - SiempreViernes 10mo ago
    I think it could be intended, there is a SI document that says something like "x /unit" is a common way to indicate the unit of a quantity, which a guy I know is using as basis for advocating for that ugly display standard.
Ethan312 10mo ago
Always good to see more competition in the inference chip space, especially from Europe. The specs look solid, but the real test will be how mature the software stack is and whether teams can get models running without a lot of friction. If they can make that part smooth, it could become a practical option for workloads that want local control.
leo_e 10mo ago
Impressive numbers on paper, but looking at their site, this feels dangerously close to vaporware. The bottleneck for inference right now isn't just raw FLOPS or even memory bandwidth—it's the compiler stack. The graveyard of AI hardware startups is filled with chips that beat NVIDIA on specs but couldn't run a standard PyTorch graph without segfaulting or requiring six months of manual kernel tuning. Until I see a dev board and a working graph compiler that accepts ONNX out of the box, this is just a very expensive CGI render.
- mg 10mo ago
  Six months of one developer tuning the kernel? That seems like not much compared to the hundreds of billions of dollars US companies currently invest into their AI stack? OpenAI pays thousands of engineers and researchers full time.
  - NaomiLehman 10mo ago
    more like 100 developers for 2 years
    - nrhrjrjrjtntbt 10mo ago
      its the new "...and tell me if the picture has a bird"
  - SilverBirch 10mo ago
    It is. The problem is latency. All these fields are moving very fast, and so it doesn't sound bad spending 6 months tuning something, but in reality what is happening is that during those 6 months the guy who built the thing you're tuning has iterated 5 more times and what you started on 6 months ago is now much much better than what you got handed 6 months ago whilst simultaneously being much worse than what that person has in their hands today. If the field you're working in is relatively static, or your performance gap is large enough it makes sense. But in most fields the performance gap is large in absolutely terms but small in temporal terms. You could make something run 10x faster, but you can't build something that will run faster than what will be state of the art in 2 months.
ano-ther 10mo ago
I don’t get the negativity. The specs look impressive. It is always good to have competition. They announced tapeout in October with planned dev boards next year. Vaporware is when things don’t appear, not when they are on their way (it takes some time for hardware). It’s also strategically important for Europe to have its own supply. The current and last US administration have both threatened to limit supply of AI chips to European countries, and China would do the same (as they have shown with Nexperia). And of course you need the software stack with it. They will have thought of that. https://vsora.com/vsora-announces-tape-out-of-game-changing-inference-chip-putting-europe-at-the-forefront-of-data-center-ai/ https://vsora.com/vsora-announces-tape-out-of-game-changing-...
- NaomiLehman 10mo ago
  im guessing the negativity is caused by bad branding
- vlorr 10mo ago
  The negativity doesn’t make much sense. The specs are strong, and the chip already taped out in October that’s a concrete milestone, not vaporware. Hardware of this class always takes months between tape-out and dev boards. Official announcement: https://vsora.com/vsora-announces-tape-out-of-game-changing-inference-chip-putting-europe-at-the-forefront-of-data-center-ai/ https://vsora.com/vsora-announces-tape-out-of-game-changing-... Multiple independent sources confirmed the tape-out: EE Times: https://www.eetimes.eu/vsora-tapes-out-ai-inference-chip-for-data-centers/ https://www.eetimes.eu/vsora-tapes-out-ai-inference-chip-for... L’Informaticien: https://www.linformaticien.com/magazine/infra/64028-vsora-met-jotunn8-en-production.html https://www.linformaticien.com/magazine/infra/64028-vsora-me... Solutions Numériques: https://www.solutions-numeriques.com/vsora-franchit-un-cap-avec-sa-puce-jotunn8-dediee-a-linference-ia/ https://www.solutions-numeriques.com/vsora-franchit-un-cap-a... There’s also an industrial manufacturing partnership with GUC: https://www.design-reuse.com/news/202529700-vsora-and-guc-partner-on-jotunn8-datacenter-ai-inference-processor/ https://www.design-reuse.com/news/202529700-vsora-and-guc-pa... Strategically, having a European AI inference chip matters. The US has already threatened export limits to Europe, and China has shown similar behavior (e.g., Nexperia). Building local supply is important. Calling this vaporware makes no sense: tape-out + published roadmap = real, not slides.
disdi 10mo ago
Esperanto tried to do the same but went out of business. https://www.esperanto.ai/products/ https://www.esperanto.ai/products/
- nickserv 10mo ago
  Looks like the design lives on. Wonder if it'll get any traction. https://www.opensourceforu.com/2025/11/ainekko-turns-esperantos-ai-silicon-into-open-hardware-for-the-world/ https://www.opensourceforu.com/2025/11/ainekko-turns-esperan...
  - disdi 10mo ago
    they have not yet opensourced the RTL design files
thevania 10mo ago
reminds me of the famous tachyum prodigy vapourware https://www.tachyum.com/ https://www.tachyum.com/
- eigenspace 10mo ago
  Well, this has already taped-out whereas the entire reason people call the tachyum prodigy vapourware is that they keep missing their target dates for tapeout and keep delaying it.
cardameu 10mo ago
I can ensure you it's not vaporware at all. silicon is running in the fab, application boards have finished the design phase, software stack validated...
- simondotau 10mo ago
  [flagged]
  - cardameu 10mo ago
    need to create an account to reply to people like you who pretend to know. I do not promise anything, I am just giving facts.
    - simondotau 10mo ago
      In this context, "giving facts" is synonymous with promising. I'm not pretending to know anything and that is exactly the point. I don't know anything. Nobody in here knows anything. It's all just claims. Words are cheap. Flashy websites are cheap. Press releases claiming tape-out are cheap. Until there is real silicon in the hands of real people free to benchmark it, it's indistinguishable from vaporware. Actually, given how much money is floating around in the AI space, I think I wouldn't be too out of line saying that it's also indistinguishable from an investor scam. I'm not saying that this is, but what independent evidence disproves this hypothesis?
- embedding-shape 10mo ago
  Guessing you work at Vsora, been there or at least know someone there? Any interest in sending out demos/sample hardware to European-based hackers who could spread the word if it ends up actually being a really nice device and developer experience? I can sacrifice myself and serve up bug reports as well :)
postexitus 10mo ago
Even if it's not vapourware, the website makes it look like one. Just look at those two graphs titled "Jotunn 8 Outperforms the Market" and "More Speed For the Bucks" (!) ; WTH?
- svantana 10mo ago
  The silly verbiage can be excused but not the graphs with completely unlabeled data points, IMO.
  - postexitus 10mo ago
    Yep that's what I mean - looks like AI slop to me.
- embedding-shape 10mo ago
  The people who build landing pages for hardware startups are usually (almost always) not the same people who design and build the actual hardware, for better or worse. A lot of the times, it's a outsourced web design agency who receive a brief, often written by business people, and finally the website is reviewed by business people, with some feedback from technical people who complain about the graphs, accuracy and so on. Then the business people say "But we need to point out we're faster than everyone else" and the engineers reply with "Sure, ok, whatever, I have actual work to do, sounds good".
  - nickserv 10mo ago
    Not just hardware startups, alas.
  - postexitus 10mo ago
    That's not an excuse. This company has more than a few handful of employees. Somebody saw those graphs without units, labels, tags and approved them. Have the CEO never opened their own webpage once and scrolled down?
    - 10mo ago
ndom91 10mo ago
I'll believe it when I see it wishing them the best! > To streamline development and shorten time-to-market, VSORA embraces industry standards: our toolchain is built on LLVM and supports common frameworks like ONNX and PyTorch, minimizing integration effort and customer cost.
qwertox 10mo ago
One has got to love the fact hat you only get more information if you submit your email address.
numbers_guy 10mo ago
Does anyone know why they brand it an "inference chip"? Is it something at the hardware level that makes is unsuitable for training, or is it simply that the toolchain for training is massively more complicated to program?
- yaantc 10mo ago
  Very simplified, AI workloads need compute and communications and compute dominates inference, while communications dominate training. Most start-ups innovate on the compute side, whereas the techno needed for state of the art communications is not common, and very low-level: plenty of analog concerns. The domain is dominated by NVidia and Broadcom today. This is why digital start-ups tend to focus on inference. They innovate on the pure digital part, which is compute, and tend to use off-the-shelf IPs for communications, so not a differentiator and likely below the leaders. But in most cases coupling a computation engine marketed for inference with state of the art communications would (in theory) open the way for training too. It's just that doing both together is a very high barrier. It's more practical to start with compute, and if successful there use this to improve the comms part in a second stage. All the more because everyone expects inference to be the biggest market too. So AI start-ups focus on inference first.
  - Fnoord 10mo ago
    They also have the 'tyr 4' [1]. It doesn't have to compete on price 1:1. Ever since Trump took office, the Europeans woke up on their dependence on USA who they no longer regard as a reliable partner. This counts for defense industry, but also for critical infrastructure, including IT. The European alternatives are expected to cost something. [1] https://vsora.com/products/tyr/ https://vsora.com/products/tyr/
- IshKebab 10mo ago
  Probably because their software only supports inference. It's relatively easy to do via ONNX. Training requires an order of magnitude more software work.
randomgermanguy 10mo ago
The fact that I have to give them an email for details just feels immediately like a B2B-scam. Hope they can figure out software, but what im seeing isn't super-promising
Hendrikto 10mo ago
> This is not just faster inference. It’s a new foundation for AI at scale. Did they generate their website with their own chips or on Nvidia hardware?
steeve 10mo ago
hey, we (ZML) happen to know them very well. they are incredible.
AIorNot 10mo ago
How does this compare to Euclyds product (another new EU AI chip company)? https://euclyd.ai/ https://euclyd.ai/
- yaantc 10mo ago
  I'm sorry I won't share much details, I don't think much is public on Vsora architecture and don't want to breach any NDA... From their web page Euclyd is a "many small cores" accelerator. Doing good compilation toolchains for these to get efficient results is a hard problem, see many comments on compilers for AI in this thread. Vsora approach is much more macroscopic, and differentiated. By this I mean I don't know anything quite like it. No sea of small cores, but several more beefy units. They're programmable, but don't look like a CPU: the HW/SW interface is at a higher level. A very hand-wavy analogy with storage would be block devices vs object storage, maybe. I'm sure more details will surface when real HW arrive.
cherryteastain 10mo ago
Hopefully they do better than UK's Graphcore who seem to be circling the drain

161 points by rdg42 10mo ago

all2 10mo ago

288GB RAM on board, and RISC V processors to enable the option for offloading inference from the host machine entirely. It sounds nice, but how much is it?

rq1 10mo ago
The next generation will include another processor to offload the inference from the RISC V processors used to offload inference from the host machine.
- ddalex 10mo ago
  The next next generation will include memory to offload memory from the on chip memory to the memory on memory (also known as SRAM cache)

pclmulqdq 10mo ago

It needs a "buy a card" link and a lot more architectural details. Tenstorrent is selling chips that are pretty weak, but will beat these guys if they don't get serious about sharing. Edit: It kind of looks like there's no silicon anywhere near production yet. Probably vaporware.

layer8 10mo ago
Tapeout apparently completed last month, dev boards in early 2026: https://www.eetimes.eu/vsora-tapes-out-ai-inference-chip-for-data-centers/ https://www.eetimes.eu/vsora-tapes-out-ai-inference-chip-for...
embedding-shape 10mo ago
Nice wave they've been able to ride if it's vaporware, considering they're been at it for five years. Any guesses to why no one else seemingly see the obvious you see?
- pclmulqdq 10mo ago
  Look at the CGI graphics and indications in their published material that all they have is a simulation. A It's all there without disclosing an anticipated release date. Even their product pages and their news page don't seem to have indications of this. Also, the 3D graphic of their chip on a circuit board is missing some obvious support pieces, so it's clearly not from a CAD model. Lots of chip startups start as this kind of vaporware, but very few of them obfuscate their chip timelines and anticipated release dates this much. 5 years is a bit long to tapeout, but not unreasonable.
  - embedding-shape 10mo ago
    > Even their product pages and their news page don't seem to have indications of this. This seems indicative enough for me, give or take a quarter or two probably, from the latest news post on their website: > VSORA is now preparing for full-scale deployment, with development boards, reference designs, and servers expected in early 2026. https://vsora.com/vsora-announces-tape-out-of-game-changing-inference-chip-putting-europe-at-the-forefront-of-data-center-ai/ https://vsora.com/vsora-announces-tape-out-of-game-changing-... Seems they have partners as well, who describe working together with a Taiwanese company as well. You never know, guess they could have gotten others to fall for their illusions too, it's not unheard of. But considering how long time something like this takes to bring to market, that they have dev-boards ready is months rather than years at least gives me enough to wait until then to judge them too harshly.
    - lukan 10mo ago
      "that they have dev-boards ready is months rather than years at least gives me enough to wait until then to judge them too harshly." So far, they just talk about it.
- simondotau 10mo ago
  The company itself doesn't have to be vaporware in order for the product, as described, to be vaporware. And even if it's not true vaporware, there's still no independent evidence of anything. The design could end up being a bust, whether because of unexpected hardware limitations, or an immature software stack, or underwhelming thermals, or sheer economics.

unit149 10mo ago

[dead]

N_Lens 10mo ago

An FP8 performance of 3200TFLOPS is impressive, could be used for training as well as inference. "Close to theory efficiency" is a bold statement. Most accelerators achieve 60-80% of theoretical peak; if they're genuinely hitting 90%+, that's impressive. Now let's see the price.

bangaladore 10mo ago

I love that the JS loads so slow on first load that it just says "The magic number: 0 /tflops"

unwind 10mo ago
It loaded fine for me, but that slash before the unit was a bit smelly. :| Just a tiny edit, but it's a rather core part of their message so they should probably notice and format it correctly before publishing.
- SiempreViernes 10mo ago
  I think it could be intended, there is a SI document that says something like "x /unit" is a common way to indicate the unit of a quantity, which a guy I know is using as basis for advocating for that ugly display standard.

Ethan312 10mo ago

Always good to see more competition in the inference chip space, especially from Europe. The specs look solid, but the real test will be how mature the software stack is and whether teams can get models running without a lot of friction. If they can make that part smooth, it could become a practical option for workloads that want local control.

leo_e 10mo ago

Impressive numbers on paper, but looking at their site, this feels dangerously close to vaporware. The bottleneck for inference right now isn't just raw FLOPS or even memory bandwidth—it's the compiler stack. The graveyard of AI hardware startups is filled with chips that beat NVIDIA on specs but couldn't run a standard PyTorch graph without segfaulting or requiring six months of manual kernel tuning. Until I see a dev board and a working graph compiler that accepts ONNX out of the box, this is just a very expensive CGI render.

mg 10mo ago
Six months of one developer tuning the kernel? That seems like not much compared to the hundreds of billions of dollars US companies currently invest into their AI stack? OpenAI pays thousands of engineers and researchers full time.
- NaomiLehman 10mo ago
  more like 100 developers for 2 years
  - nrhrjrjrjtntbt 10mo ago
    its the new "...and tell me if the picture has a bird"
- SilverBirch 10mo ago
  It is. The problem is latency. All these fields are moving very fast, and so it doesn't sound bad spending 6 months tuning something, but in reality what is happening is that during those 6 months the guy who built the thing you're tuning has iterated 5 more times and what you started on 6 months ago is now much much better than what you got handed 6 months ago whilst simultaneously being much worse than what that person has in their hands today. If the field you're working in is relatively static, or your performance gap is large enough it makes sense. But in most fields the performance gap is large in absolutely terms but small in temporal terms. You could make something run 10x faster, but you can't build something that will run faster than what will be state of the art in 2 months.

ano-ther 10mo ago

I don’t get the negativity. The specs look impressive. It is always good to have competition. They announced tapeout in October with planned dev boards next year. Vaporware is when things don’t appear, not when they are on their way (it takes some time for hardware). It’s also strategically important for Europe to have its own supply. The current and last US administration have both threatened to limit supply of AI chips to European countries, and China would do the same (as they have shown with Nexperia). And of course you need the software stack with it. They will have thought of that. https://vsora.com/vsora-announces-tape-out-of-game-changing-inference-chip-putting-europe-at-the-forefront-of-data-center-ai/ https://vsora.com/vsora-announces-tape-out-of-game-changing-...

NaomiLehman 10mo ago
im guessing the negativity is caused by bad branding
vlorr 10mo ago
The negativity doesn’t make much sense. The specs are strong, and the chip already taped out in October that’s a concrete milestone, not vaporware. Hardware of this class always takes months between tape-out and dev boards. Official announcement: https://vsora.com/vsora-announces-tape-out-of-game-changing-inference-chip-putting-europe-at-the-forefront-of-data-center-ai/ https://vsora.com/vsora-announces-tape-out-of-game-changing-... Multiple independent sources confirmed the tape-out: EE Times: https://www.eetimes.eu/vsora-tapes-out-ai-inference-chip-for-data-centers/ https://www.eetimes.eu/vsora-tapes-out-ai-inference-chip-for... L’Informaticien: https://www.linformaticien.com/magazine/infra/64028-vsora-met-jotunn8-en-production.html https://www.linformaticien.com/magazine/infra/64028-vsora-me... Solutions Numériques: https://www.solutions-numeriques.com/vsora-franchit-un-cap-avec-sa-puce-jotunn8-dediee-a-linference-ia/ https://www.solutions-numeriques.com/vsora-franchit-un-cap-a... There’s also an industrial manufacturing partnership with GUC: https://www.design-reuse.com/news/202529700-vsora-and-guc-partner-on-jotunn8-datacenter-ai-inference-processor/ https://www.design-reuse.com/news/202529700-vsora-and-guc-pa... Strategically, having a European AI inference chip matters. The US has already threatened export limits to Europe, and China has shown similar behavior (e.g., Nexperia). Building local supply is important. Calling this vaporware makes no sense: tape-out + published roadmap = real, not slides.

disdi 10mo ago

Esperanto tried to do the same but went out of business. https://www.esperanto.ai/products/ https://www.esperanto.ai/products/

nickserv 10mo ago
Looks like the design lives on. Wonder if it'll get any traction. https://www.opensourceforu.com/2025/11/ainekko-turns-esperantos-ai-silicon-into-open-hardware-for-the-world/ https://www.opensourceforu.com/2025/11/ainekko-turns-esperan...
- disdi 10mo ago
  they have not yet opensourced the RTL design files

thevania 10mo ago

reminds me of the famous tachyum prodigy vapourware https://www.tachyum.com/ https://www.tachyum.com/

eigenspace 10mo ago
Well, this has already taped-out whereas the entire reason people call the tachyum prodigy vapourware is that they keep missing their target dates for tapeout and keep delaying it.

cardameu 10mo ago

I can ensure you it's not vaporware at all. silicon is running in the fab, application boards have finished the design phase, software stack validated...

simondotau 10mo ago
[flagged]
- cardameu 10mo ago
  need to create an account to reply to people like you who pretend to know. I do not promise anything, I am just giving facts.
  - simondotau 10mo ago
    In this context, "giving facts" is synonymous with promising. I'm not pretending to know anything and that is exactly the point. I don't know anything. Nobody in here knows anything. It's all just claims. Words are cheap. Flashy websites are cheap. Press releases claiming tape-out are cheap. Until there is real silicon in the hands of real people free to benchmark it, it's indistinguishable from vaporware. Actually, given how much money is floating around in the AI space, I think I wouldn't be too out of line saying that it's also indistinguishable from an investor scam. I'm not saying that this is, but what independent evidence disproves this hypothesis?
embedding-shape 10mo ago
Guessing you work at Vsora, been there or at least know someone there? Any interest in sending out demos/sample hardware to European-based hackers who could spread the word if it ends up actually being a really nice device and developer experience? I can sacrifice myself and serve up bug reports as well :)

postexitus 10mo ago

Even if it's not vapourware, the website makes it look like one. Just look at those two graphs titled "Jotunn 8 Outperforms the Market" and "More Speed For the Bucks" (!) ; WTH?

svantana 10mo ago
The silly verbiage can be excused but not the graphs with completely unlabeled data points, IMO.
- postexitus 10mo ago
  Yep that's what I mean - looks like AI slop to me.
embedding-shape 10mo ago
The people who build landing pages for hardware startups are usually (almost always) not the same people who design and build the actual hardware, for better or worse. A lot of the times, it's a outsourced web design agency who receive a brief, often written by business people, and finally the website is reviewed by business people, with some feedback from technical people who complain about the graphs, accuracy and so on. Then the business people say "But we need to point out we're faster than everyone else" and the engineers reply with "Sure, ok, whatever, I have actual work to do, sounds good".
- nickserv 10mo ago
  Not just hardware startups, alas.
- postexitus 10mo ago
  That's not an excuse. This company has more than a few handful of employees. Somebody saw those graphs without units, labels, tags and approved them. Have the CEO never opened their own webpage once and scrolled down?
  - 10mo ago

ndom91 10mo ago

I'll believe it when I see it wishing them the best! > To streamline development and shorten time-to-market, VSORA embraces industry standards: our toolchain is built on LLVM and supports common frameworks like ONNX and PyTorch, minimizing integration effort and customer cost.

qwertox 10mo ago

One has got to love the fact hat you only get more information if you submit your email address.

numbers_guy 10mo ago

Does anyone know why they brand it an "inference chip"? Is it something at the hardware level that makes is unsuitable for training, or is it simply that the toolchain for training is massively more complicated to program?

yaantc 10mo ago
Very simplified, AI workloads need compute and communications and compute dominates inference, while communications dominate training. Most start-ups innovate on the compute side, whereas the techno needed for state of the art communications is not common, and very low-level: plenty of analog concerns. The domain is dominated by NVidia and Broadcom today. This is why digital start-ups tend to focus on inference. They innovate on the pure digital part, which is compute, and tend to use off-the-shelf IPs for communications, so not a differentiator and likely below the leaders. But in most cases coupling a computation engine marketed for inference with state of the art communications would (in theory) open the way for training too. It's just that doing both together is a very high barrier. It's more practical to start with compute, and if successful there use this to improve the comms part in a second stage. All the more because everyone expects inference to be the biggest market too. So AI start-ups focus on inference first.
- Fnoord 10mo ago
  They also have the 'tyr 4' [1]. It doesn't have to compete on price 1:1. Ever since Trump took office, the Europeans woke up on their dependence on USA who they no longer regard as a reliable partner. This counts for defense industry, but also for critical infrastructure, including IT. The European alternatives are expected to cost something. [1] https://vsora.com/products/tyr/ https://vsora.com/products/tyr/
IshKebab 10mo ago
Probably because their software only supports inference. It's relatively easy to do via ONNX. Training requires an order of magnitude more software work.

randomgermanguy 10mo ago

The fact that I have to give them an email for details just feels immediately like a B2B-scam. Hope they can figure out software, but what im seeing isn't super-promising

Hendrikto 10mo ago

> This is not just faster inference. It’s a new foundation for AI at scale. Did they generate their website with their own chips or on Nvidia hardware?

steeve 10mo ago

hey, we (ZML) happen to know them very well. they are incredible.

AIorNot 10mo ago

How does this compare to Euclyds product (another new EU AI chip company)? https://euclyd.ai/ https://euclyd.ai/

yaantc 10mo ago
I'm sorry I won't share much details, I don't think much is public on Vsora architecture and don't want to breach any NDA... From their web page Euclyd is a "many small cores" accelerator. Doing good compilation toolchains for these to get efficient results is a hard problem, see many comments on compilers for AI in this thread. Vsora approach is much more macroscopic, and differentiated. By this I mean I don't know anything quite like it. No sea of small cores, but several more beefy units. They're programmable, but don't look like a CPU: the HW/SW interface is at a higher level. A very hand-wavy analogy with storage would be block devices vs object storage, maybe. I'm sure more details will surface when real HW arrive.

cherryteastain 10mo ago

Hopefully they do better than UK's Graphcore who seem to be circling the drain