@NVIDIAAP profile picture
NVIDIA Asia Pacific @NVIDIAAP

The official Twitter channel for all NVIDIA news, products and events in Asia Pacific. Follow us on Facebook: facebook.com/NVIDIAAP

Joined August 2017
  • Posts 5,687
  • Following 17
  • Followers 9,433
3,784 Photos and videos
Increase inference performance by up to 15x without sacrificing responsiveness. DFlash, an open source lightweight block diffusion model designed for speculative decoding, delivers up to 15x higher throughput on NVIDIA Blackwell while maintaining the same user interactivity target. Instead of drafting tokens one at a time, it proposes a whole block in a single pass for the main model to verify in parallel. Adoption is drop-in with support in @lmsysorg SGLang, TensorRT-LLM, and @vllm_project.
2
5
31
1,208
Read the full deep dive here: nvda.ws/4fofpGj

Boost Inference Performance up to 15x on NVIDIA Blackwell Using DFlash Speculative Decoding |...

As AI systems move from single-turn interactions to coordinated multiagent workflows, low-latency inference becomes increasingly important. Autoregressive LLMs generate tokens sequentially…

developer.nvidia.com
4
144
NVIDIA Asia Pacific reposted
NVIDIA Newsroom
 @nvidianewsroom
Jun 30
NVIDIA BioNeMo Agent Toolkit is now integrated with Anthropic’s Claude Science. Life sciences organizations increasingly need AI that goes beyond chat — agents that can reason over scientific questions and act on them using real computational tools. NVIDIA BioNeMo Agent Toolkit gives Claude the domain-specific scientific context to select the right tool, prepare valid inputs, execute workflows and interpret results. blogs.nvidia.com/blog/claude…

NVIDIA BioNeMo Agent Toolkit Brings Accelerated AI to Life Sciences Researchers in Claude Science

Anthropic’s Claude Science, an AI workbench for science research, lets scientists use agents to build workflows harnessing NVIDIA BioNeMo models, libraries and accelerated workflows.

blogs.nvidia.com
9
46
232
10,066
What does it actually take to move a telecom network from automated to truly autonomous? Our latest tech blog outlines the key building blocks of a telco autonomy platform that let agents move safely through problem-solution loops at higher levels of autonomy. 🔗 Read now to learn how to build autonomous networks with agentic AI: nvda.ws/4p9QB9s
1
1
22
670
NVIDIA Asia Pacific reposted
NVIDIA GTC
 @NVIDIAGTC
Jun 30
Get the best price for #NVIDIAGTC Berlin conference passes today: ✅ GTC alumni get the biggest savings ✅ Early bird pricing is open to everyone ✅ All discounted pricing ends July 15 🔗nvda.ws/4ashiQs
3
9
34
21,178
NVIDIA Asia Pacific reposted
NVIDIA GTC
 @NVIDIAGTC
Jun 29
Today’s agentic AI is built on models that can reason—and now, with projects like OpenClaw, we’re starting to feel AI not just give us answers, but actually do work for us as autonomous, long‑running, self‑evolving agents. Watch the Agentic AI 101 session now. nvda.ws/4eDGHYv
0:37
3
3
21
1,668
⏰ Last chance to apply! Applications for the Inception Grand Challenge 2026 now close on July 5. Dive into real-world enterprise problem statements from confirmed enterprises, including DFI, Cooler Master, and VNG, with more coming soon. Don't miss your opportunity to engage with leading enterprises, refine pilot-ready AI solutions, and showcase your innovation on a regional stage. Join now 👉 nvda.ws/4awCzsh #NVIDIAInception
2
15
912
Owning the network used to be the competitive advantage. In the AI era, it's the starting point. Join IDC's Ahmad Latif Ali, Singtel's Manoj Kumar and NVIDIA's Chris Penrose as they discuss how telco operators across Asia Pacific are shifting from generative AI experimentation to real-time AI inferencing, distributed AI architectures and new AI-powered revenue opportunities. Watch the executive briefing on-demand: nvda.ws/3RcbXGu
2
17
551
Revolut built a transaction foundation model – accelerated by the NVIDIA full-stack platform on Nebius – to improve fraud detection, product recommendations, and other use cases in financial services. The results: 📈 2.3x better credit risk accuracy ⚡ Up to 5x higher training throughput
0:51
3
15
57
35,187
Learn how accelerated computing is redefining infrastructure efficiency in banking and payments: nvda.ws/4eV0X8e

Revolut's Transaction Foundation Model is 3-5x Faster on NVIDIA H100 GPUs

See how Revolut uses PRAGMA, a behavioral model, to unify fraud, credit, and marketing systems. They achieved major gains in accuracy (2.3x better credit risk, 65% more fraud cases caught) using...

nvidia.com
1
4
431
Your AI agent shouldn't need the cloud to think. NVIDIA NemoClaw is an open-source stack for building secure, always-on AI agents that run entirely on your own hardware — no external services, no data leaving the device. Built on NVIDIA DGX Spark, it combines sandboxed execution, real-time policy controls, and local inference with a 120B Nemotron model to deliver enterprise-grade AI with full data residency. nvda.ws/4gic4dL
10
21
153
6,453
For enterprises across APAC where data sovereignty and compliance aren't optional — this is the on-premises AI stack built for you. For developers: full deployment guide, Jupyter notebook, and open-source code available now.
8
376
💡 What breakthroughs have APAC enterprises achieved with #NVIDIADGX? Across the region, leaders like @HonHai_Foxconn, @MediaTek, @official_naver, and @NTTData are utilizing DGX SuperPOD to build agentic AI factories, scale intelligence from edge to cloud, and deliver sovereign AI. #DecadeOfDGX 🔗 Learn more about how customers use the DGX platform as the blueprint for the modern AI factory: nvda.ws/4avsX0S
0:55
1
4
33
988
Deploying RAG in a notebook is one thing, but taking it to production is another level. Join this full-day, instructor-led NVIDIA workshop to learn how to build and scale production-ready RAG pipelines using NVIDIA NIM and Kubernetes. Build it live with an expert and leave with skills you can bring straight back to your team. Register now: nvda.ws/4wf1w3E
2
6
34
745
NVIDIA Asia Pacific reposted
NVIDIA AI Infrastructure
 @NVIDIAAIInfra
Jun 17
What does it actually take to connect hundreds of thousands of GPUs with zero jitter and predictable latency? Gilad Shainer, SVP of Networking at NVIDIA, walks through the full stack of networking innovations enabling gigascale AI: Quantum-X InfiniBand, Spectrum-X Ethernet with co-packaged optics, and SuperNICs. Watch the full breakdown from #NVIDIAGTC Taipei on demand ➡️ nvda.ws/4uHafuA
10
29
226
12,921
📣 @AnthropicAI Claude models are now generally available in @Microsoft Foundry, running on the NVIDIA GB300 NVL72 platform on Azure with NVIDIA Quantum-X800 InfiniBand networking. For enterprises building the next generation of agentic AI, this means more powerful autonomous and domain-specific agents, backed by the inference performance and lowest cost per token.
2
11
63
1,679
🔗 Learn more: nvda.ws/4gNAf3W

Claude Meets Blackwell Ultra: Anthropic’s Models Now Run on NVIDIA GB300 in Azure

Now generally available in Microsoft Foundry, Claude on NVIDIA GB300 Blackwell Ultra gives Azure-native enterprises a new foundation for building autonomous and domain-specific AI agents.

blogs.nvidia.com
167
The future of computer graphics is here, and it's built by AI. ✨ Don't miss the NVIDIA Research Keynote at SIGGRAPH 2026 for live demos and research examples showcasing breakthrough neural rendering, world models, and simulation methods across creative tools, industrial design, robotics, and autonomous systems.
2
3
35
868
Learn more: nvda.ws/4fazcd9

Keynote Presentations - SIGGRAPH 2026

The Premier Conference & Exhibition on Computer Graphics & Interactive Techniques. 53 years of SIGGRAPH and join us in Los Angeles starting 19 July.

s2026.siggraph.org
3
150
🎁 Join the webinar and unlock your next step: Learn how to build a safer autonomous AI agent with NVIDIA Brev credits. The first 200 attendees who join early and stay until the end of the webinar will have a chance to receive free NVIDIA Brev credits to complete the hands-on course "How to Build a Safer Autonomous Agent using OpenClaw" Continue your learning after the webinar with guided exercises covering agent security, model selection, and deploying long-running autonomous agents using NVIDIA Nemotron, NemoClaw, OpenClaw, and NVIDIA NeMo. 📅 July 9 | 2:00 p.m. SGT | 4:00 p.m. AEST 👉 Secure your spot: nvda.ws/4uGsGzy
2
9
37
978
Load more