sometimes it's the GPUs and sometimes it's infuriating like this because you have the GPUs and something dumb is causing a bottleneck
AI HARDWARE
-
NVLink vs Cerebras Fabric: Interconnect Speed Comparison
By
–
nvlink is measured in terabytes/s cerebras fabric is measured in petabytes/s
-
OpenAI GPT-OSS released due to Chinese labs pressure
By
–
i rate gpt-oss, it's just that openai wouldn't have released that model without opensource Chinese Labs putting so much pressure on them and yes, Buy a GPU 🙂
-
Chinese XPENG Creates Most Human-Like Humanoid Robot Ever
By
–
This is an insane leap! The most human-like humanoid robot in history!
— Amitav Bhattacharjee (@bamitav) 7 novembre 2025
Designed by the Chinese company #XPENG
pic.twitter.com/RtcYunTeOL#humanoidrobotics #humanoid #AI #Robot #humanoidrobotics @enilev @efipm @FutureStacked @sulefati7 @Fisher85MThis is an insane leap! The most human-like humanoid robot in history! Designed by the Chinese company #XPENG #humanoidrobotics #humanoid #AI #Robot #humanoidrobotics @enilev @efipm @FutureStacked @sulefati7 @Fisher85M
-
Robot Dancer Showcased at China Import Expo Event
By
–
Robot dancer steals the spotlight at China Import Expo https://
search.app/fj2UE #Robotics #Robot #humanoidrobot #AI @smaksked @sulefati7 @Shi4Tech @SpirosMargaris @mvollmer1 -
Tesla Optimus Humanoid Robot Demonstrates Dancing Capabilities
By
–
Optimus can dance too🕺
— Amitav Bhattacharjee (@bamitav) 7 novembre 2025
pic.twitter.com/6ZhOjnGPq1#Tesla #optimus #Tesla #humanoid #AI@CurieuxExplorer @CyberRobooo @Eli_Krumova @Nicochan33Optimus can dance too #Tesla #optimus #Tesla #humanoid #AI @CurieuxExplorer @CyberRobooo @Eli_Krumova @Nicochan33
-
Tesla Optimus Hand Unveiled at Annual Shareholder Meeting
By
–
Optimus hand on display at Tesla Annual Shareholder Meeting!
— Amitav Bhattacharjee (@bamitav) 7 novembre 2025
pic.twitter.com/0dOiVIY4QU#teslabot #optimus #Tesla #Robotics #AI @mary_gambara @Fisher85M @AlbertoEMachado @mvollmer1Optimus hand on display at Tesla Annual Shareholder Meeting! #teslabot #optimus #Tesla #Robotics #AI @mary_gambara @Fisher85M @AlbertoEMachado @mvollmer1
-
Google Launches 7th Generation TPU Ironwood with Improved Performance
By
–
Our 7th generation TPU (Ironwood) is becoming generally available for Cloud TPU customers. It's got greatly improved performance and efficiency over previous generations! 🎉 https://t.co/shW4srbI1E
— Jeff Dean (@JeffDean) 6 novembre 2025Our 7th generation TPU (Ironwood) is becoming generally available for Cloud TPU customers. It's got greatly improved performance and efficiency over previous generations!
-
Blackwell versus Cerebras: AI Hardware Processor Comparison Analysis
By
–
More in our blog: https://
cerebras.ai/blog/blackwell
-vs-cerebras
… -
Cerebras Outperforms Nvidia H100 with 3000 Tokens Per Second
By
–
Cerebras beats Nvidia H100 but can it beat Blackwell? Blackwell inference endpoints are finally out and it’s fast. It runs GPT-OSS-120B at ~700 tokens/s, leapfrogging H100 and Groq. Cerebras clocked in at 3,000 TPS – still #1. Looking forward to Rubin!