BRN Discussion Ongoing
- Thread starter TechGirl
- Start date
smoothsailing18
Regular
FF
While IBM proving out AKD1000 is huge don't forget this is just one more proof of commercial application that Brainchip is heir too:
Case Study: Jetson + Akida vs. Conventional AI
At the National Security Collaboration Center inside UT San Antonio's San Pedro I building, the AI Cowboys tested a hybrid system combining the NVIDIA Jetson Orin AGX with a BrainChip Akida PCIe Board.
Traditional Jetson-only workload: Continuous anomaly detection on video streams at 120 watts.
Jetson + Akida neuromorphic hybrid: The same workload, the same accuracy—at just 9 watts.
That’s a 90% reduction in energy. To put it in perspective, that’s like shutting off 1,000 household lightbulbs every single hour.
Sustainability
If deployed globally, neuromorphic AI could reduce energy consumption enough to eliminate CO₂ emissions equivalent to those of millions of cars. It’s not just green tech, it’s survival tech for industries under ESG pressure.
Accessibility
Because these systems consume minimal power, they can operate in off-grid or low-infrastructure regions. That means farmers monitoring crops, doctors using portable diagnostic kits, or disaster teams running autonomous drones, all without relying on cloud data centers or stable electricity
While IBM proving out AKD1000 is huge don't forget this is just one more proof of commercial application that Brainchip is heir too:
Case Study: Jetson + Akida vs. Conventional AI
At the National Security Collaboration Center inside UT San Antonio's San Pedro I building, the AI Cowboys tested a hybrid system combining the NVIDIA Jetson Orin AGX with a BrainChip Akida PCIe Board.
Traditional Jetson-only workload: Continuous anomaly detection on video streams at 120 watts.
Jetson + Akida neuromorphic hybrid: The same workload, the same accuracy—at just 9 watts.
That’s a 90% reduction in energy. To put it in perspective, that’s like shutting off 1,000 household lightbulbs every single hour.
Sustainability
If deployed globally, neuromorphic AI could reduce energy consumption enough to eliminate CO₂ emissions equivalent to those of millions of cars. It’s not just green tech, it’s survival tech for industries under ESG pressure.
Accessibility
Because these systems consume minimal power, they can operate in off-grid or low-infrastructure regions. That means farmers monitoring crops, doctors using portable diagnostic kits, or disaster teams running autonomous drones, all without relying on cloud data centers or stable electricity
TheDrooben
Pretty Pretty Pretty Pretty Good
Larry got his last parcel just before close......400k now. Predicting a sell-off tomorrow now. Sorry about that
Happy as Larry
Shorts see no momentum, so they keep leaning on it. No pressure, no urgency…nothing that suggests a real covering squeeze is building.
Looks like they already did lol
Two new Patents that have been publicly released in 2026 that appear on the European site that, mention us
in the artwork, yes, I know, so what I hear some say, the point that you are missing is that we are the leading
commercial offering verses research chips (maybes one day) IBM True North and Intel Loihi.
Being mentioned (in dispatches) so what, having the words Brainchip, Akida means we are at least "out there"
Check them out, one new company, one regular.........
Some may appreciate that our company is getting more and more exposure, some may not.
God Bless All................Tech.
in the artwork, yes, I know, so what I hear some say, the point that you are missing is that we are the leading
commercial offering verses research chips (maybes one day) IBM True North and Intel Loihi.
Being mentioned (in dispatches) so what, having the words Brainchip, Akida means we are at least "out there"
Check them out, one new company, one regular.........
Some may appreciate that our company is getting more and more exposure, some may not.
God Bless All................Tech.
Its highly unlikely that Kevin Johnson is talking BS.
His position at IBM:
Field CTO – HPC, AI, LLM & Quantum Computing | Principal HPC Cloud Technical Specialist at IBM | Symphony • GPFS • LSF
He is a Field CTO, Technical Specialist.
He gets paid to know what he is talking about.
Would he risk his reputation and stick his neck out if he was not sure about what he is posting?
The market does what it does.
Jacob Kiraz
Regular
Ha ha another dude that just sits around and wont be vocal, deep down your hurtingYeah…. That’s right… send him where he belongs to….
Fullmoonfever
Top 20
I see Michael has updated some of his GitHub for the TENNs project he's doing.
#github #repository #multiagent #agenticai #workflow #decisionmaking #smalllanguagemodels #slms #cnn #rnn #tenns #autonomous #uav #uas #cuas #antidrone #management #platforms | Michael Liubchenko
Some updates in my Github repos: 1️⃣ Agentic API performance measurements. 🎯 Agentic AI performance comparison for Java, Python, Go, Rust implementations. 2️⃣ Model Compilers. 🎯 This project focuses on evaluating Apache TVM and OpenXLA as ML compilers for deploying two TENNs models (a Large...www.linkedin.com
View attachment 94624
View attachment 94625
I hope Michael Liubchenko means “decent” rather than “descent”, though…
Has some benchmarking info and great to see he is able to run on CPU, GPU and Apple.
EEG-Edge-Model/BENCHMARK_GUIDE.md at main · mlyu2010/EEG-Edge-Model
Develop, create train and quantize a comprehensive AI models based on TENN (Temporal Event-based neural networks or State Space Recurrent Models) using 2D images and 1D convolutions that are optimi...
github.com
Performance Comparison
Expected Performance (TENN EEG Model)
| Device | Framework | Avg Time | Throughput | Notes |
|---|---|---|---|---|
| CPU | ONNX | ~6ms | ~160/s | Optimized for CPU |
| CPU | PyTorch | ~60ms | ~15/s | Less optimized |
| CUDA | PyTorch | ~2ms | ~500/s | Requires GPU |
| MPS | PyTorch | ~3ms | ~330/s | Apple Silicon |
You miss understand what I’m saying… it’s not about his statement… of course he knows what he is talking about … it’s more about what we are always interpreting from that what we are reading… we always tend to see deals and rising shares and so on.
Instead, the reality always show us falling share price and after a week a complete silence of the topics we was hyping
Edit: you remember someone got a call from management because he shared his excitement about our tech… he also knew what he is talking about! And we got also excited before the call
Last edited:
HopalongPetrovski
I'm Spartacus!
Yes Fur. Are you seriously just understanding this?
Here, we are predominantly positive around the possabilities about BrainChip.
Over on the crapper they have been infested with posters who are mainly negative.
In years past, before this site was created, we were all fighting it out over there.
BRN was the most popular and patronised thread there for years.
It used to be ok over there with a fairly even mix, but predominately positive viewpoints, until the shorts discovered the stock.
Particularly after the Mercedes spike and subsequent fall it gradually got nastier and nastier and so Zeebot saw an opportunity and created this place.
Here we are willing to entertain the full spectrum of thought on the stock but when it is just continuous and repetitive down ramping for its own sake we tend to jack up and piss them off.
Here, we are predominantly positive around the possabilities about BrainChip.
Over on the crapper they have been infested with posters who are mainly negative.
In years past, before this site was created, we were all fighting it out over there.
BRN was the most popular and patronised thread there for years.
It used to be ok over there with a fairly even mix, but predominately positive viewpoints, until the shorts discovered the stock.
Particularly after the Mercedes spike and subsequent fall it gradually got nastier and nastier and so Zeebot saw an opportunity and created this place.
Here we are willing to entertain the full spectrum of thought on the stock but when it is just continuous and repetitive down ramping for its own sake we tend to jack up and piss them off.
Me too mainly positiv.. I’m a buyer (unfortunately)
but I’m also pissed… some people don’t know the difference between a downramper and a shareholder who is worried…
Pom down under
Top 20
Waiting for a few off the regular downrampers to manage to downramp this someway or another.
Sign Up | LinkedIn
500 million+ members | Manage your professional identity. Build and engage with your professional network. Access knowledge, insights and opportunities.www.linkedin.com
View attachment 94815 View attachment 94816
AI OverviewNice find!
Well, we know that Frontgrade Gaisler in Dec 2024 licensed Akida IP for rad hardened, space grade system on chip .............
I doubt very much any other neuromorphic chip is that far!
Frontgrade Gaisler Licenses BrainChip Akida IP to Power AI in Space
Frontgrade Gaisler partners with BrainChip to integrate Akida IP, enabling advanced AI chip deployment for reliable, low-power processing in space missions.brainchip.com
Based on recent reports, SpaceX is heavily upgrading the processing power of its Starlink satellites with advanced, adaptive, and AI-focused hardware to improve data handling, rather than specifically using "neuromorphic" (brain-inspired) chips.
Here are the key details regarding the technology powering the new generation of Starlink satellites:
- AMD Versal AI Core Integration: Starlink satellites are utilizing AMD Versal AI Core adaptive system-on-chip (SoC) CPUs. These chips are designed for high-performance AI inference and adaptive processing, allowing for faster, more efficient data processing directly in orbit.
- "Mini Laser" Technology: SpaceX has developed and is deploying "mini laser terminals" that allow for a 25Gbps connection between satellites, serving as a mesh network. These terminals are being integrated into other satellite networks (like Muon Space) to provide real-time, low-latency connectivity.
- Edge Computing in Space: The integration of these advanced chips aims to enable "in-orbit edge processing," which allows for near-real-time data fusion and AI inference, reducing the need to send all raw data back to Earth.
- Power Efficiency: The new technology is designed to operate under strict power constraints, with some AI-enabled edge processing solutions aiming to operate at less than 1 Watt, which is crucial for satellite functionality.
- AI Training Potential: As of February 2026, updated privacy policies indicate that Starlink data may be used to train artificial intelligence models.
Similar threads
