Search

AIopenmind
AIopenmind @AIopenmind ·
Jensen Huang sostiene che Nvidia entri nella fase dell’“inference inflection”: il focus passa dall’addestramento all’uso operativo dell’AI, con un’opportunità di ricavi da 1.000 miliardi . #Nvidia #AI #Inference #JensenHuang #GTC2026 aiopenmind.it/ArtificialInte…
Nvidia e l’“inference inflection”: la nuova corsa dell’IA
Nvidia e l’“inference inflection”: la nuova corsa dell’IA

Jensen Huang rilancia la traiettoria di Nvidia e sostiene che il mercato stia entrando nell’era dell’“inference inflection”, cioè nella fase in cui il valore dell’intelligenza artificiale si sposta…

From aiopenmind.it
13
SwiftInference.ai
SwiftInference.ai @swiftinference ·
AI inference is no longer experimental in telecoms - it's in the operational core. From real-time fraud detection to predictive network maintenance, the infrastructure decisions operators make now will define their competitive position for years.swiftinference.ai/blog/how-ai-in… #inference
How AI Inference Is Transforming Telecommunications in 2026 — SwiftInference Blog

AI inference is moving from the data centre to the network edge, fundamentally reshaping how telecoms operators manage infrastructure, reduce churn, and deliver service quality. Here is what the...

From swiftinference.ai
3
Groookounet
Groookounet @groookounet ·
Replying to @groookounet
2/10 Le choc de mars 2026 : Nvidia vient de racheter l'unité d'inférence de Groq pour 20 Md$. 💰 Pourquoi ? Parce que les puces classiques saturent. Groq apporte la vitesse "temps réel" dont les agents IA ont besoin. La guerre de l'inférence est déclarée. ⚔️ #inference
1
15
Sarbjeet Johal
Sarbjeet Johal @sarbjeetjohal ·
Disaggregated #inference is changing how teams design AI architectures on @Kubernetes — splitting prefill and decode into distinct services with different resource profiles and scaling needs. #kubecon @SantoshYadavDev @SaiyamPathak @dhinchcliffe @nyike @dvellante @furrier @rseroter @kaslinfields @virtualized6ix @IsForAt @NVIDIAAIDev @NVIDIADC @NVIDIAAI @NVIDIAAIDev
NVIDIA Data Center NVIDIA Data Center @NVIDIADC ·
💡 Disaggregated LLM inference is changing how teams design AI architectures on Kubernetes — splitting prefill and decode into distinct services with different resource profiles and scaling needs. Learn how to: ✅ Separate prefill and decode for better GPU utilization andUse gang scheduling, hierarchical gang scheduling, and topology-aware placement ✅ Express multi-role inference pipelines with APIs like LeaderWorkerSet and NVIDIA Grove 🔗 Read the tech blognvda.ws/4lK2nWluv
4
553
Deva
Deva @DevaCodeX ·
Replying to @DevaCodeX
2️⃣ Gimlet Labs raised $80M to crack AI's inference bottleneck. Their multi-silicon cloud splits workloads across CPUs, GPUs & custom chips for 3-10x faster performance. Already at 8-figure revenue. techcrunch.com/2026/03/23/sta… #AI #Inference
Startup Gimlet Labs is solving the AI inference bottleneck in a surprisingly elegant way | TechCr...

Gimlet Labs just raised an $80 million Series A for tech that lets AI run across NVIDIA, AMD, Intel, ARM, Cerebras and d-Matrix chips, simultaneously.

From techcrunch.com
1
29
Harry He
Harry He @harryheisme ·
Office Hour Time! I've been flooded with messages asking: "I just opened the SGLang repo, How do I not waste my first day?" Join this Office Hour and get your answers directly! #sglang #inference #inferenceframework #lmsys #AI
LMSYS Org LMSYS Org @lmsysorg ·
📣 New to SGLang? No problem — Our Office Hours have you covered 👌 This week's session is built for beginners: "New to SGLang: What I Learned & What I Wish I Knew on Day 1." 👉Alex Nails (@@alxnails), MTS at @@radixark, is sharing what it's actually like to onboard into SGLang — hher, what took some time to click, and his ideas on what could be better. Join us for the mental model walkthrough for SGLang, and an open discussion on making the dev and learning experience better. 📅 March 25 | 6:00 PM PST Register on Luluma.com/87xexrbgbWjM
56