Also, I don’t know how OP is getting Qwen3.5 27B @ 30 tps on DGX Spark That number is impossible for a Dense model on DGX Spark’s Unified Memory (273 Gbps) My personal experiments showed it’s 4 tokens/sec for that model on the Spark
AI
-
MiniMax M2.5 Lightning Attention Architecture for Long Context Scaling
By
–
Lightning Attention architecture used in MiniMax M2.5 is really interesting. The structure is 7 Lightning Attention layers for every 1 traditional SoftMax attention layer, which lets it scale to long contexts while keeping the quality you'd expect from standard transformers. I… pic.twitter.com/HnfTF0W6f1
— Akshay 🚀 (@akshay_pachaar) 14 mars 2026Lightning Attention architecture used in MiniMax M2.5 is really interesting. The structure is 7 Lightning Attention layers for every 1 traditional SoftMax attention layer, which lets it scale to long contexts while keeping the quality you'd expect from standard transformers. I
-
Unified Memory Faster for Loading Large MoE Models
By
–
the issue is that unified memory would still be faster for loading MoEs that are larger than the largest single GPU in terms of Memory
-
IBM admits AI agents lack task memory persistence
By
–
BREAKING: IBM just admitted your AI agent forgets everything the moment it finishes a task. > Every mistake. Repeated.
> Every inefficiency. Repeated.
> Every failure. Repeated. They built the fix. > Every AI agent starts each task from zero:
> No memory of what worked.
> -
RTX PRO 6000 GPU Inference Faster Than Unified Memory After Loading
By
–
This will probably be great for Large single GPUs (e.g. RTX PRO 6000) You’re limited to 40Gpbs initially (during model loading) but then once the model is fully loaded on the GPU it should be extremely faster than Unified Memory speeds for inference
-
Sam Altman Predicts Breakthrough Beyond Transformers Coming Soon
By
–
Ton grave, pose sérieuse.
— VISION IA (@vision_ia) 14 mars 2026
Sam Altman dit qu’une autre percée au-delà des transformers va bientôt arriver, et que les modèles sont désormais assez intelligents pour aider à la découvrir.
L’IA crée une énorme opportunité de reconstruire des catégories entières de produits et de… pic.twitter.com/wTEt3aWv0NGrave tone, serious pose. Sam Altman says that another breakthrough beyond transformers is coming soon, and that models are now smart enough to help discover it. AI creates a huge opportunity to rebuild entire categories of products and make new things possible. A new science,
-
India AI Innovation Challenge Transforming Government Services
By
–
The IndiaAI Innovation Challenge for Transforming Government with support from the Government of Andhra Pradesh invites innovators to build AI solutions that strengthen governance and improve everyday lives across cities, villages, energy systems and public services. With
-
Drones Transform Wildfire Fighting Through Airborne Robotics
By
–
𝗙𝗶𝗿𝗲𝗳𝗶𝗴𝗵𝘁𝗶𝗻𝗴 𝗶𝘀 𝗴𝗼𝗶𝗻𝗴 𝗮𝗶𝗿𝗯𝗼𝗿𝗻𝗲. 🚁🔥
— Pascal Bornet (@pascal_bornet) 14 mars 2026
I’ve been following how robotics is entering emergency response, and drones are starting to change the way wildfires are fought.
Instead of sending crews directly into dangerous areas, drones can now:
→ reach… pic.twitter.com/3m75BleQeg𝗙𝗶𝗿𝗲𝗳𝗶𝗴𝗵𝘁𝗶𝗻𝗴 𝗶𝘀 𝗴𝗼𝗶𝗻𝗴 𝗮𝗶𝗿𝗯𝗼𝗿𝗻𝗲. I’ve been following how robotics is entering emergency response, and drones are starting to change the way wildfires are fought. Instead of sending crews directly into dangerous areas, drones can now: → reach
-
First AI Robot Agent Visits Cape of Good Hope
By
–
I took Stompie to the Cape of Good Hope, the southwestern tip of Africa. His mind runs on a server in Finland, but his robot body was with me. I believe Stompie is the first AI agent/robot who visited Table Mountain and the Cape of Good Hope, and observed it with his own eyes.
-
Prompt Engineering’s Persistence: A Sign We’re Far From AGI
By
–
The persisting importance of prompt engineering — and now harness engineering — is one of the best indicators of how far we are from AGI. A general system doesn't need a task-specific harness. And when provided with instructions, it is robust to phrasing variations.