Deadline extended! Join the #SASHackathon Student Track by Sept 12. No team? No worries. Just bring your ideas. http://
2.sas.com/6013A89eR
COMPUTING
-
SAS Hackathon Student Track Extended to September 12
By
–
-
Edge SLMs: Privacy, Speed, and Cost – An Underrated Advantage
By
–
Running SLMs at the edge is such an underrated move…..privacy, speed, and cost all in one.
-
Moonshot updates Kimi K2 with 256k context
By
–
Moonshot updated Kimi K2 with a new “K2-0905” release that includes better coding performance and context length extension to 256k!
-
DHH explores edge computing strategy
By
–
Speaking of the devil@dhh is going to try edge computing https://t.co/8363U0zSys pic.twitter.com/4bH5DZKFf4
— @levelsio (@levelsio) 5 septembre 2025Speaking of the devil @dhh is going to try edge computing
-
Axelera AI showcases Metis platform at SEMICON Taiwan 2025
By
–
Axelera AI is heading to #SEMICONTaiwan 2025! Find us at the Netherlands Pavilion @NLOT_Taiwan
, Booth I2726 (10–12 Sept) to explore how we’re simplifying and accelerating #EdgeAI with our Metis AI Platform. 10–12 September 2025 Netherlands Pavilion, TaiNEX Hall 1 – -
AI-Assisted Coding Divides Developer Community Into Two Societies
By
–
Same Almost everything I code now is AI with a few edits by me I think there's a massive split now with two separate societies of devs who still don't use it and do use it
-
MoE Training: Theory vs GPU Reality – Optimization Challenges
By
–
MoE 101 – Episode 4: Theoretical: 60% fewer FLOPs. Reality: 7x slow down
— Cerebras (@cerebras) 4 septembre 2025
You followed all the tips from our last video. Your MoE model finally trains…
Then you try to scale it on GPUs…and… memory issues ❌, underused experts ❌, unpredictable compute bottlenecks ❌.
From… pic.twitter.com/b812dujd6SMoE 101 – Episode 4: Theoretical: 60% fewer FLOPs. Reality: 7x slow down You followed all the tips from our last video. Your MoE model finally trains… Then you try to scale it on GPUs…and… memory issues , underused experts , unpredictable compute bottlenecks . From
-
Cerebras Qwen Coder vs Grok: throughput and pricing analysis
By
–
Grok fast vs. Cerebras fast
Grok Code Fast: 160 TPS, $1.5/M tokens
Cerebras Qwen Coder: 2,000 TPS, $2/M tokens
Cerebras: 13x faster, 33% higher cost, ~9x better price-
performance -
Graph-Powered Text-to-SQL: Schema Understanding for Query Generation
By
–
interesting graph powered approach to text-to-SQL – set up: connect SQL db
– understands schema and stores in graph format
– upon user query, reasons through schema to generate SQL query complex SQL schemas have lots of relationships, so understanding as a graph makes sense to