BRN Discussion Ongoing

[Rach2512]

MosChip® on LinkedIn: #ai #automotivetech #ocr

Optical Character Recognition (OCR) is revolutionizing the automotive industry! From license plate recognition to driver assistance, OCR enhances safety &…

www.linkedin.com

 

[Hrdwk]

This might sound stupid, But I wonder if they would consider doing what DeepSeek did to ChatGPT to DeepSeek with Akida?

 

[uiux]

https://github.com/Brainchip-Inc/akida_examples/releases/tag/2.11.0-doc-1

Latest version of Akida docs states "Aligned with FPGA 1562(2-node)/1563(6-node)/1532(ViT)"


Indicates three upcoming FPGAs:

• FPGA 1562(2-node)
• 1563(6-node)
• 1532(ViT)

 

[Fullmoonfever]

Haven't had much time this week & just skim reading posts where I can.

Just ran an ASX Broker data on BRN to go with a couple I saved the other week to see who's doing what.

Not gone through properly myself yet but thought I'd post them here fwiw as about to shutdown lappy.

One I hadn't seen before was in the first date range, Instinet Aust (part of the Nomura Group it appears) who bought $440k worth for someone(s).

Newest date range to oldest covering ~ 3 days ago (T+3 data) back to 1/12.

 

[Guzzi62]

https://github.com/Brainchip-Inc/akida_examples/releases/tag/2.11.0-doc-1

Latest version of Akida docs states "Aligned with FPGA 1562(2-node)/1563(6-node)/1532(ViT)"


Indicates three upcoming FPGAs:

• FPGA 1562(2-node)
• 1563(6-node)
• 1532(ViT)

Care to explain what that mean?

Software update right?

But what that mean practically I have no idea, guess you can compare it to all your apps on your phone getting updated?

 

[uiux]

Care to explain what that mean?

Software update right?

But what that mean practically I have no idea, guess you can compare it you all your apps on your phone getting updated?

They have updated the software to support features of the three FPGAs

Eg. ChatGPT explains

That note in the GitHub release document likely refers to the specific FPGA (Field-Programmable Gate Array) versions that the release aligns with. This suggests that the software or model update in the release is optimized for or compatible with these particular FPGA configurations.

Here's a breakdown of the numbers:

• FPGA 1562 (2-node) → This FPGA version is designed for a 2-node architecture.
• FPGA 1563 (6-node) → This FPGA version is configured for a 6-node system.
• FPGA 1532 (ViT) → This FPGA version is optimized for ViT (Vision Transformer) models.

 

[Fullmoonfever]

They have updated the software to support features of the three FPGAs

Eg. ChatGPT explains

That note in the GitHub release document likely refers to the specific FPGA (Field-Programmable Gate Array) versions that the release aligns with. This suggests that the software or model update in the release is optimized for or compatible with these particular FPGA configurations.

Here's a breakdown of the numbers:

• FPGA 1562 (2-node) → This FPGA version is designed for a 2-node architecture.
• FPGA 1563 (6-node) → This FPGA version is configured for a 6-node system.
• FPGA 1532 (ViT) → This FPGA version is optimized for ViT (Vision Transformer) models.

@uiux

You probs understand GitHub better and wondering if you have any thoughts on a previous post of mine from a little while ago?

Post in thread 'BRN Discussion Ongoing' https://thestockexchange.com.au/threads/brn-discussion-ongoing.1/post-444388

 

[uiux]

Question for our more technically knowledgable like @Diogenese or @chapman89

I did a search on a keyword and found @Sirod69 had been looking at it some time ago (though she sadly hasn't been seen for a few months?) as this evening I found what I thought could be an interesting entry on Github regarding the open source Zephyr RTOS.

Brainchip-India lodged a Documentation Issue a couple of days ago.

Now, is Brainchip-India "our" Indian team?.....can't think of any other Brainchip in India that might be playing with RTOS (Real Time Operating Systems) for SOCs.

Anyway, what caught my eye was the issue being related to the below and in particular "Nordic" SOCs.

Is it conceivable that our team are either playing with the Nordic NRF5 series of SOCs of their own accord or potentially (& preferably) at the request of a client using one of the NRF5 series SOCs?

Or is it a non (SNN) event...pun intended.

doc: Documentation issue in 'develop/flash_debug/nordic_segger' · Issue #83191 · zephyrproject-rtos/zephyr

Describe the bug << Please describe the issue here >> << Looks like the tool mentioned "nrfjprog" to flash elf onto nRF board is no longer available and the official nRF came-up with a common "nrfu...

github.com

View attachment 74568

Some background on Zephyr

Zephyr Project

The Zephyr Project is a Linux Foundation hosted Collaboration Project.

zephyrproject.org

View attachment 74573

View attachment 74574

View attachment 74575

View attachment 74576

View attachment 74577

Some background on the Nordic nRF5 series of SOCs of which a lot end uses appear in our wheelhouse.

Search

Nordic Semiconductor

www.nordicsemi.com

nRF5x SoC Overview

nRF5x SoC Overview The Nordic Semiconductor nRF5x System on Chip (SoCs) are ideal for ultra-low power (ULP) and cost effective short-range wireless solutions. Typical applications for the nRF5x SoC…

embeddedcentric.com

View attachment 74578

Could be either / or


Maybe working on a prototype

 

[Diogenese]

https://github.com/Brainchip-Inc/akida_examples/releases/tag/2.11.0-doc-1

Latest version of Akida docs states "Aligned with FPGA 1562(2-node)/1563(6-node)/1532(ViT)"


Indicates three upcoming FPGAs:

• FPGA 1562(2-node)
• 1563(6-node)
• 1532(ViT)

Thanks ui,

This is a peek into the BRN labs.

FPGAs are like the old breadboard circuit assembly to test new configurations in silicon without having to go to full tapeout/wafer production.

2 nodes = 8 NPUs.

6 nodes = 24 NPUs.

ViT = Vision Transformer - this is an option for Akida 2.

I assume these are Akida 2, and from what I've seen of what they have published, each node includes TENNs.

2 nodes is lightweight, but considerably larger than Pico.

6 node would be quite powerful.

ViT is for tracking moving objects. As Vit is an optional extra for Akida2, and Akida 2 has TENNs, and TENNs can process video, I wonder if there is any overlap between the functions?

Oh to be a fly on the wall ...

 

[IloveLamp]

CES 2025

Check back frequently for updates, as we’ll be adding new information throughout CES 2024 and beyond.

brainchip.com

Alf Kuchenbuch on LinkedIn: Here's our review for CES 2025: https://lnkd.in/dvQSMKyQ

Here's our review for CES 2025: https://lnkd.in/dvQSMKyQ

www.linkedin.com

 

[IloveLamp]

https://www.embedded.com/neuromorphic-computing-the-new-frontier-of-computer-science-inspired-by-the-human-brain/

Frank Schirrmeister on LinkedIn: Neuromorphic Computing: The New Frontier of Computer Science Inspired by…

As we are working through the #AI craziness and an industry, I have been following various new computing architectures. I had an "aha-moment" a couple of years…

www.linkedin.com

 

[AusEire]

New Podcast

 

[MDhere]

Morning fellow brners, happy Friday!
We should be receiving the top 20 shareholders notice today? I checked back on previous notices and quite a number of top 20 increased their shareholdings so will be interesting to compare when this one comes out. O to be a top 20...
Whatever our individual investments are we are part of this growing company and soon imo Brainchip will be thriving just like Arm, Nividia , Amazon and the rest of them that grew into giants
Feeling positive on the last day of January. Looking forward to seeing the Top 20 chart. Bring it on, boomidty boom boom

 

[Humble Genius]

Morning fellow brners, happy Friday!
We should be receiving the top 20 shareholders notice today? I checked back on previous notices and quite a number of top 20 increased their shareholdings so will be interesting to compare when this one comes out. O to be a top 20...
Whatever our individual investments are we are part of this growing company and soon imo Brainchip will be thriving just like Arm, Nividia , Amazon and the rest of them that grew into giants
Feeling positive on the last day of January. Looking forward to seeing the Top 20 chart. Bring it on, boomidty boom boom

What would be even more positive today is if we received an Ann to the tune of the following

Sydney – 31 January 2025 – BrainChip Holdings Ltd (ASX: BRN, OTCQX: BRCHF, BCHPY) (“BrainChip”, or the “Company”), the world’s first commercial producer of neuromorphic artificial intelligence IP, announces today that it has signed multiple year production contracts with Lockhead Martin, Raytheon, Airbus, Boeing, Rolls Royce, Spacex, Collins Aerospace, etc. Sean Hehir, Brainchip CEO, commented on these announcements, "I'm glad these contracts have come to fruition, otherwise my arse would've got reamed big time at the next AGM".

 

[IloveLamp]

What would be even more positive today is if we received an Ann to the tune of the following

Sydney – 31 January 2025 – BrainChip Holdings Ltd (ASX: BRN, OTCQX: BRCHF, BCHPY) (“BrainChip”, or the “Company”), the world’s first commercial producer of neuromorphic artificial intelligence IP, announces today that it has signed multiple year production contracts with Lockhead Martin, Raytheon, Airbus, Boeing, Rolls Royce, Spacex, Collins Aerospace, etc. Sean Hehir, Brainchip CEO, commented on these announcements, "I'm glad these contracts have come to fruition, otherwise my arse would've got reamed big time at the next AGM".

 

[Bloodsy]

Another great podcast! This one a bit more technical but absolutely fascinating hearing about radar. Our mate Dr Tapson is very impressive, as time goes on we are seeing more and more the caliber of our employees.

My favorite teaser

"2025 is going to be a really exciting year for us, because we have - we can speak about it broad terms, we have some very exciting radar projects going ahead with some prime contractors in the US aviation industry" - Dr Tapson, vice pres engineering.

19'10 time stamp in the latest podcast if you'd like to hear it for yourself

 

[Taproot]

@uiux
From CES, day 3 wrap up.
Mentions Akida 2 FPGA
https://www.linkedin.com/company/brainchip-holdings-limited/posts

BrainChip BrainChip 19,288 followers19,288 followers2w • 2 weeks ago

BrainChip at hashtag#CES2025: Day 3 Recap

Day 3 brought exciting milestones as CMO Steve Brightfield appeared live on "The Gadget Guy" discussing neuromorphic computing's critical role in advancing hashtag#EdgeAI. All Things AI podcast included conversations with Jeff Bier, founder of the Edge AI and Vision Alliance, and Ramkishor Korada, CEO of Vedya.ai, exploring the challenges of Edge AI product development and how expert solutions are reshaping the field.

Our demo suite was buzzing with activity as visitors from leading companies explored our technology. Live demos of audio denoising, Edge LLM and Akida 2 FPGA platform impressed with their real-world applications.

BrainChip's CEO, Sean Hehir, participated in video sessions with Taiwan’s Industrial Technology Research Institute (ITRI)(工業技術研究院, 工研院) and discussions UK-based ipXchange, extending our global reach and showcasing the impact of BrainChip’s innovations worldwide.

We’re thrilled with the momentum so far and look forward to an exciting Day 4!
More: https://lnkd.in/gSYtwfgi

 

[buena suerte :-)]

The good Doctor is an absolute ️


View attachment 76954


DeepSeek R1's breakthrough results are a boon for BrainChip and the extreme-edge AI industry. Here's how it will impact inference at the extreme edge:

DeepSeek has shattered the "No one ever got fired for going with IBM" mentality, liberating customers from the behemoths' models. This paves the way for a Cambrian explosion of specialized LLMs delivered by smaller companies. It's akin to breaking the 4-minute mile – now that it's been done, we're poised to witness an explosion of innovation. This is excellent news for smaller companies like BrainChip.

Now imagine AI models running on whispers of energy, delivering powerful performance without massive data centers or cloud connectivity. This is the promise of extreme-edge AI, which is BrainChip's sweet spot.

While DeepSeek is an impressive transformer based model, TENNs technology, based on State Space Models (SSMs), currently offers a superior solution for extreme-edge applications:

1. No explosive KV cache: Unlike traditional transformer models that rely on rapidly expanding key-value (KV) caches, TENNs sidesteps this issue with fixed memory requirements, regardless of input length. This fundamental property enables LLMs at the extreme edge.

2. Competitive Performancereliminary experiments pitting the DeepSeek 1.5B model against TENNs 1.2B model in various tasks, such as trip planning and simple programming, showed comparable or slightly better results for TENNs.

3. Extremely Low Training Costs: BrainChip's focus on small specialized models means training costs are less than a premium economy flight from Los Angeles to Sydney.

DeepSeek's success highlights areas where TENNs can be further enhanced. We can leverage many of the tricks learned from DeepSeek to improve their TENNs LLM even more.

The future of extreme-edge AI is bright, with DeepSeek demonstrating that small companies can compete effectively in this space.

Great read ...Nice work @Tothemoon24 ....

Bring it home BRN...

"While DeepSeek is an impressive transformer based model, TENNs technology, based on State Space Models (SSMs), currently offers a superior solution for extreme-edge applications:"

 

[buena suerte :-)]

https://github.com/Brainchip-Inc/akida_examples/releases/tag/2.11.0-doc-1

Latest version of Akida docs states "Aligned with FPGA 1562(2-node)/1563(6-node)/1532(ViT)"


Indicates three upcoming FPGAs:

• FPGA 1562(2-node)
• 1563(6-node)
• 1532(ViT)

Great to see you back with your very informative posts @uiux ...

Cheers