[GUEST ACCESS MODE: Data is scrambled or limited to provide examples. Make requests using your API key to unlock full data. Check https://lunarcrush.ai/auth for authentication information.]

@NVIDIAAIDev "ICYMI: AI Agent Hackathon was 🔥 But don't worry if you missed out. The same powerful workshop framework used by participants is now a launchable for you to build agents. Dive in 👉"
@NVIDIAAIDev on X 2025-07-29 18:00:09 UTC 78.5K followers, 1410 engagements

"🎉Congratulations to @Microsoft for the new Phi-4-mini-flash-reasoning model trained on NVIDIA H100 and A100 GPUs. This latest edition to the Phi family provides developers with a new model optimized for high-throughput and low-latency reasoning in resource-constrained environments. Bring your data and try out demos on the multimodal playground for Phi on the NVIDIA API Catalog ➡ 📷 The first plot shows average inference latency as a function of generation length while the second plot illustrates how inference latency varies with throughput. Both experiments were conducted using the vLLM"
@NVIDIAAIDev on X 2025-07-18 23:14:41 UTC 78.5K followers, 11.3K engagements

"🔐 AI can no longer rely on assumed trust. Learn how cryptographic model signing on NGC builds trust into agentic systems that reason plan and act. Tech blog ➡"
@NVIDIAAIDev on X 2025-07-28 18:05:01 UTC 78.5K followers, 4141 engagements

"Prototype fine-tune and deploy with ease. 🙌 NVIDIA DGX Spark your own personal AI supercomputer. ✅ X petaFLOP ✅ 128GB of coherent unified system memory ✅ ConnectX-7 Smart NIC ✅ Up to 4TB Storage ✅ 150mm L x 150mm W x 50.5mm H Reserve yours now ➡"
@NVIDIAAIDev on X 2025-07-01 17:30:16 UTC 78.5K followers, 24.9K engagements

"📣 Announcing the release of OpenReasoning-Nemotron: a suite of reasoning-capable LLMs which have been distilled from the DeepSeek R1 0528 671B model. Trained on a massive high-quality dataset distilled from the new DeepSeek R1 0528 our new 7B 14B and 32B models achieve SOTA perf on a wide range of reasoning benchmarks for their respective sizes in the domain of mathematics science and code. The models are available on @huggingface🤗:"
@NVIDIAAIDev on X 2025-07-18 18:50:29 UTC 78.5K followers, 54.5K engagements

"🚦Hitting performance walls in pandas We ran X example workflows that slowed or crawled on large datasetsthen reran the exact same code on GPU with cudf.pandas. TLDR: ✅ 18M rows of stock prices 2040x faster with time-based rolling windows ✅ 8GB job postings CSV up to 30x faster on string ops + merges ✅ 7.3M geospatial points interactive dashboards that stay interactive Same pandas code. More speed. 🏎🏁 🔗 Check out the blog for links to Colab demos and setup guides:"
@NVIDIAAIDev on X 2025-07-27 21:57:13 UTC 78.5K followers, 3463 engagements

"Your 3D Gaussian Splats look great but physics They've been a flop. Discover how to do deformable simulations on 3DGS at #NVIDIAResearch's NVIDIA Kaolin & Warp workshop at #SIGGRAPH2025 👉"
@NVIDIAAIDev on X 2025-07-25 18:00:58 UTC 78.5K followers, 4973 engagements

"📣 Announcing Llama Nemotron Super v1.5 📣 This release pushes the boundaries of reasoning model capabilities at the weight class of the model and is ready to power agentic applications from individual developers all the way to enterprise applications. 📈 The Llama Nemotron Super v1.5 achieves leading reasoning accuracies for science math code and agentic tasks while delivering up to 3x higher throughput. Try it on or download from @huggingface: 🤗 Tech blog:"
@NVIDIAAIDev on X 2025-07-26 00:02:00 UTC 78.5K followers, 19.7K engagements

"Were excited to share that 🥇Llama Nemotron Super v1.5 -- our latest open reasoning model -- is now #1 on the @ArtificialAnlys Intelligence Index - a leaderboard that spans advanced math science and agentic tasks in the 70B open model category. Super v1.5 is trained with high-quality reasoning synthetic data generated from models like Qwen3-235B and DeepSeek R1. It delivers state-of-the-art accuracy and throughput running on a single H100. Key features: 🎯 Leading accuracy on multi-step reasoning math coding and function-calling 🏗 Post-trained using RPO DPO and RLVR across 26M+ synthetic"
@NVIDIAAIDev on X 2025-07-29 18:45:12 UTC 78.5K followers, 3156 engagements

"Were excited to share that 🥇Llama Nemotron Super 49B v1.5 -- our latest open reasoning model -- is now #1 on the @ArtificialAnlys Intelligence Index - a leaderboard that spans advanced math science and agentic tasks in the 70B open model category. Llama Nemotron Super 49B v1.5 is trained with high-quality reasoning synthetic data generated from models like Qwen3-235B and DeepSeek R1. It delivers state-of-the-art accuracy and throughput running on a single H100. Key features: 🎯 Leading accuracy on multi-step reasoning math coding and function-calling 🏗 Post-trained using RPO DPO and RLVR"
@NVIDIAAIDev on X 2025-07-29 19:35:53 UTC 78.5K followers, 29.2K engagements

"🏎 Just one line of code can accelerate FLUX-1 Dev by up to 2.4X. ⚡ Using FP8 quantization and its seamless LoRA support Torch-TensorRT unlocks peak GPU performance with maximum flexibility. Technical deep dive:"
@NVIDIAAIDev on X 2025-07-25 17:00:01 UTC 78.5K followers, 2210 engagements

"@ArtificialAnlys 📝 See our Tech Blog for more details:"
@NVIDIAAIDev on X 2025-07-29 21:33:38 UTC 78.5K followers, 1133 engagements

"📣 Introducing NVIDIA AI Dev Weekly 📰 Compiled every Friday we're delivering technical deep dives trainings events and key updates for developers all in one read. Stay up to date on the latest in the world of generative AI data science accelerated computing and more👇"
@NVIDIAAIDev on X 2025-07-25 20:04:50 UTC 78.5K followers, 3157 engagements

"How can medical researchers efficiently extract detailed and accurate data from complex free-text medical records 🩺 📊 A multidisciplinary team @UTSWMedCenter developed an LLM pipeline to automatically extract and standardize information from pathology reports with near perfect accuracy"
@NVIDIAAIDev on X 2025-07-29 16:15:01 UTC 78.5K followers, 1177 engagements