Create

Sign in to ReadmeX

or

How we handle your data: Privacy Policy

ReadmeX

Your AI reading companion

What would you like to catch up on?

Pick a briefing, or ask your own question.

Your community and people briefings stay personal to you.

NVIDIA

AI overview

Sign in and the AI will write an overview from our coverage.

Headlines · 1493

  1. Nvidia reports $58.1bn in investments and $118bn in credit guarantees

    Nvidia spent $58.1bn buying shares in other companies in the nine months ending 25 October, including stakes in OpenAI and Anthropic, according to its quarterly results cited by Fortune and The Next Web. The company also reported $32bn in future commitments to neocloud providers and $118bn in credit guarantees, including $105bn for the PORTS-Pike data centre in Ohio, which OpenAI plans to lease for 20 years. Nvidia said some guarantees take effect only after a customer has secured financing. The stakes generated $1.8bn in net gains during the quarter.

    The Next Web · 🔥 0
  2. Castrol launches CORE liquid-cooling solution for AI data centers

    Castrol launched Castrol CORE in China at the Data Center Standards Conference, combining design support, cooling technologies, monitoring hardware and lifecycle services for liquid-cooled AI data centers. The company also issued a white paper on operating and maintaining cold-plate liquid-cooling systems, while reporting initial orders and shipments for a monitoring sensor and simulated heating device. Castrol says its cooling technologies have been tested with Intel and Nvidia, but the supplied report provides no independent test or shipment data.

    智东西 · 🔥 0
  3. Supermicro shareholder lawsuit targets alleged China GPU sales

    Shareholders’ civil lawsuit alleges that former Supermicro executive Cenly Chen knew about alleged illicit China sales and controlled allocations of Nvidia racks to customers. Supermicro denied that Chen managed its Nvidia relationship or decided GPU distribution, while sources cited by Tom’s Hardware said former Nvidia sales chief Jay Puri, who is married to Chen, oversaw chip allocations to server makers. Three Supermicro employees, including co-founder Yih-Shyan “Wally” Liaw, were arrested over an alleged $2.5 billion AI GPU smuggling scheme.

    Tom’s Hardware · 🔥 0
  4. As AI Spreads, Corporate Advantage Shifts to Leadership Judgment

    TMTPost argues that AI is taking over increasingly standardized work such as retrieval, calculation, drafting and checking, making access to computing and information less differentiating. It compares Nvidia, Apple and Microsoft, then draws on management theory to argue that leaders’ problem framing, resource allocation, conflict handling and long-term objectives may matter more as technology becomes widely available. The analysis presents possible mechanisms, not proof that a particular leadership style determines business or wealth outcomes.

    钛媒体 · 🔥 0
  5. Nvidia’s interim CEO says TSMC is supporting the AI-chip ramp

    Interim CEO Colette Kress said TSMC executives visited Nvidia’s offices this week and were fully supporting the company, according to cnBeta. Nvidia raised its next-fiscal-year revenue-growth guidance from 70% to 75%, while reporting strong Blackwell shipments and continued Rubin ramp-up; Kress also said Rubin generated $20 billion in revenue during its first quarter of volume shipments.

    cnBeta · 🔥 0
  6. Nvidia Tightens Financing Guarantees for Neocloud Customers

    Nvidia is becoming more selective about financing support for emerging cloud customers, and interim CEO Colette Kress said guarantees generally take effect only after a data center’s land, power and shell are ready, according to a syndicated report citing The Wall Street Journal and Nvidia’s third-quarter earnings call. The report says this narrower support could raise neocloud borrowing costs and complicate data-center construction. It also noted sharp share-price declines for CoreWeave and Nebius following Jensen Huang’s departure, but did not establish that the guarantee policy caused those declines. Nvidia reported third-quarter revenue of $115.8 billion and data-center revenue of $107.6 billion, both above expectations, and supply commitments of $390 billion at quarter-end.

    虎嗅 AI · 🔥 0
  7. IMF Chief Warns of Further AI Market Correction and Debt Risk

    IMF Chief Kristalina Georgieva said AI-related markets may undergo further correction because investor enthusiasm is excessive, although no large-scale adjustment has occurred yet. IT之家 reported that circular AI financing totals about $1 trillion, roughly two-thirds through debt, raising financial-stability concerns as the financing structure shifts toward borrowing. Georgieva also said the IMF conservatively estimates AI could add 0.5 percentage points to global economic growth.

    Bloomberg Technology · 🔥 0
  8. Musk says Anthropic seeks 10 GW more SpaceX compute

    Elon Musk said Anthropic’s highest-probability outcome was to rent an additional 10 GW of compute from SpaceX, citing what Anthropic has told him. Wccftech reported that this conflicts with earlier reports that Anthropic executives said the compute arrangement would not be extended beyond Q2 2027. Musk also identified power generation at the power-plant level as the main constraint on SpaceX’s compute expansion, while Wccftech reported that Starbase had received permits for 1,456 MW of on-site gas turbines.

    Wccftech · 🔥 0
  9. China’s chip investment surge meets a harder commercialization test

    China’s semiconductor industry received 331.1 billion yuan across 1,475 investment events in the first three quarters of 2026, according to CVSource statistics cited by TMTPost. The analysis finds that companies are at different stages: ChangXin and Yangtze Memory have reached process upgrades and broader product portfolios, while newer AI-compute chips still need testing, software adaptation, customer validation, and mass-production preparation. It also examines how memory makers and AI accelerators are competing for the next generation of mobile, server, and AI-system demand.

    钛媒体 · 🔥 0
  10. Anthropic’s Record IPO Highlights AI Infrastructure Financing Risks

    Anthropic began trading on Nasdaq on November 19 at $83 per share, raising about $81.3 billion through its public offering while a concurrent $26 billion private placement brought total financing to roughly $107.3 billion, according to TMTPost and 钛媒体. The private placement drew major chip and cloud suppliers, while filings cited roughly $900.9 billion in computing and infrastructure commitments, highlighting how investors, customers, lenders and vendors are becoming tightly connected. 虎嗅 AI and 36氪 AI also examined Amazon’s valuation gains, employee wealth estimates and the financing risks created by long-duration commitments.

    钛媒体 · 🔥 7
  11. llm-d 1.0 adds production inference features for agent workloads

    llm-d 1.0 adds multi-tenant resource management, request-priority scheduling, high availability, rolling upgrades and context-aware routing for agent workflows. An IBM Research benchmark deployed an approximately 753-billion-parameter model through llm-d, supporting up to 3,000 concurrent coding agents at peak output throughput above 6.6 million tokens per minute. Red Hat AI 3.5 also added governance capabilities, while IBM and Red Hat jointly disclosed that Lightwell Clearinghouse had reached general availability.

    InfoQ 中文 AI&大模型 · 🔥 0
  12. Open-source OCR models cut million-page digitization cost

    Daniel van Strien of Hugging Face benchmarked eight open-source AI OCR models on 850 pages of historical books and archival documents using the same Nvidia RTX PRO 6000 GPU. At equal image sizes, RolmOCR cost about $0.51 per 1,000 pages, while LightOnOCR-3-0.8B cost just under $0.10, putting GPU rental for one million pages at $65 to $130 depending on the material. The newest model was at least as accurate as the oldest on book pages, though some models hallucinated text on blank pages or repeatedly produced the same content.

    The Decoder · 🔥 0
  13. Noitom Robotics Tops The Information’s 2026 Asia Startup List

    A supplied report says Noitom Robotics ranked first in the Asia section of The Information’s “The 50 Most Promising Startups 2026,” released on November 11. The report attributes to the list a valuation above $500 million, more than $50 million in equity financing and annual revenue above $10 million; it says these financial figures were collected directly from founders and investors. It also reports that Noitom launched its ModalityNet data platform and three flagship dataset products in June 2026. The report says the company does not manufacture robots and instead supplies standardized multimodal training data for physical-AI development.

    雷峰网 AI · 🔥 0
  14. Nvidia moves pay and equity grants forward after Huang’s exit

    Nvidia brought its annual pay review and additional stock grants forward to December at the company’s first all-hands meeting since Jensen Huang resigned. Interim CEO Colette Kress announced the change while board chair Harvey Jones discussed the CEO search, and leadership reiterated that the board did not plan to restore Huang as CEO. Co-founder Chris Malachowsky also urged employees to protect Nvidia’s founding values, which TechNews reported in a more critical account than The Next Web.

    The Next Web · 🔥 0
  15. Hennessy says a Jensen Huang-caliber Nvidia successor is possible

    Alphabet chairman John Hennessy advised Nvidia’s board to find another outstanding CEO, saying a successor as accomplished as Jensen Huang would be challenging but possible. He said no preferred candidate had been identified, cited difficulty finding an external successor, and expressed hope for an internal promotion; Nvidia did not announce a successor.

    cnBeta · 🔥 0
  16. Hennessy Says Finding Nvidia’s Next CEO Is Challenging

    Alphabet Chairman John Hennessy said in an interview with The Seoul Economic Daily that finding a successor to Nvidia’s former CEO Jensen Huang would be challenging but not impossible. When asked whether a particularly suitable candidate had been identified, he said there was no specific candidate. He advised the board to find another excellent CEO and raised both the possibility of recruiting someone from outside Nvidia and promoting someone from within.

    IT之家 AI · 🔥 0
  17. Anthropic raises record $107.3B in IPO and private placement before Nasdaq debut

    Anthropic priced its upsized IPO at $83 per share, above the marketed range of $70 to $78. The IPO and a private placement raised a combined record $107.3 billion. Seven companies, including Google, Nvidia and Microsoft, agreed to buy $26 billion of stock in the private placement at the IPO price. Shares were scheduled to begin trading on Nasdaq under the ticker ANT. Gray-market indications pointed to a roughly 20% to 27% opening gain, with estimates varying by venue and instrument.

    The Next Web · 🔥 0
  18. AI-Linked Tech Stocks Lose Lead in Rare Sector Rotation

    TechNews reports that capital is moving away from AI-linked technology shares toward defensive sectors such as healthcare and consumer staples, a shift described as unusually large for U.S. equities. Over the 100 days through November 10, S&P 500 healthcare stocks reportedly outperformed information-technology stocks by more than 33%, while Taiwan’s market fell more than 1,900 points intraday on November 19. TechNews also reports a valuation reversal, with technology stocks trading at 16.9 times forward earnings versus about 19 times for healthcare and nearly 25 times for consumer staples.

    科技新报 · 🔥 0
  19. AI-Led Tech Selloff Drives Rotation Into Defensive Stocks

    科技新报 reports that global capital is rotating away from AI-led technology shares and into defensive sectors such as healthcare and consumer staples. It cites a 100-day S&P 500 performance gap of more than 33% between healthcare and information technology through November 10, alongside a technology-sector forward P/E of 16.9. The report also says some strategists see oversold technology shares as a potential opportunity, while emphasizing that the selloff does not mean the AI industry has no future.

    科技新报 · 🔥 0
  20. AVC Details Liquid-Cooling Roadmap for High-Power AI Chips

    AVC said its existing water-cooling plates can handle about 8,000 watts as chip thermal design power moves above 6,000 watts, while it is developing a 10-kilowatt-class two-phase solution. The company also reported that its diamond-copper approach completed proof of concept and drew inquiries from NVIDIA and ASIC customers, but customers are still evaluating it and manufacturing and post-soldering reliability remain unresolved. AVC said NVIDIA plans to use a microchannel lid in Rubin Ultra and that it may participate in the design.

    科技新报 · 🔥 0
  21. NVIDIA Tightens Purchase-Guarantee Conditions for AI Neocloud Providers

    IT之家 AI, citing The Verge, reported that NVIDIA has tightened the conditions for purchase guarantees offered to AI infrastructure providers. On an earnings call, interim CEO Colette Kress said NVIDIA would screen partners and that guarantees for neocloud providers would take effect only after financing, data-center construction, compute deployment and NVIDIA’s acceptance. The guarantees would cease to apply if construction were delayed or impeded.

    IT之家 AI · 🔥 0
  22. Nvidia Turmoil Spurs Interest in Alternative AI Chips

    Microsoft executive Judson Althoff said Nvidia’s leadership turmoil strengthened customers’ interest in AMD chips, cloud providers’ in-house processors and other alternatives, though most customers are still waiting for the situation to settle. Microsoft plans to launch Maia 300 in 2027 and, according to The Information as cited by TechNews, discussed reserving production capacity for more than 300,000 chips with TSMC; no completed reservation or measured customer adoption was confirmed.

    科技新报 · 🔥 0
  23. NVIDIA SEED Cuts Prefill Work and Compresses Hybrid LLM State

    NVIDIA researchers proposed SEED, a method that uses shallower layers and lightweight emitters to write memories needed for later generation while retaining full-depth computation during decoding. On Qwen3.8-Flash-Next, parameter-dependent prefill computation fell about 25%, from 12.1 to 9.1 GFLOP per non-final input token. SEED also stores recurrent state in a compressed representation, reducing Qwen3.8-Flash-Next's per-request state from 113.2 MB to 29.2 MB immediately after compression and averaging 43.4 MB over the compression cycle. The report describes tests on five hybrid models and an arXiv preprint. In GB300 GPU and SGLang tests replaying coding-agent sessions, NVIDIA reported 16%–42% higher peak throughput per GPU; gains varied by deployment and workload.

    机器之心 · 🔥 0
  24. Atlas brings world-model camera shifts to photography workflows

    World Labs’ Atlas world model can generate new views from one or more reference images after users specify a virtual camera position and angle, according to 36氪’s account of the vendor demonstration. The report says Atlas is available through a web application and developer interface, with film and visual effects identified as an application area; it also explores how world models could assist autonomous photography hardware.

    36氪 人工智能 · 🔥 0
  25. Voltage Glitching Unlocks Up to 80GB on NVIDIA CMP 170HX

    Hardware researcher Kate used voltage fault injection to rewrite protected fuse settings on NVIDIA CMP 170HX mining cards, exposing more HBM2e memory and 120–126 working streaming multiprocessors. Nine of ten tested cards reached 80GB, while one exposed only 64GB because some memory was functional. In one ProjectPhysX FP16 test, a modified card reached 85.9 TFLOPS versus 72.5 TFLOPS for a retail A100, but ECC, MIG and other features remained untested.

    IT之家 AI · 🔥 0
  26. Kalanick Urges Huang’s Return and Backs a Conditional Nvidia Strike

    Uber co-founder Travis Kalanick said at the TomorrowX summit in Austin that Jensen Huang should return as Nvidia CEO. Kalanick said he would support or participate in an employee strike seeking Huang’s reinstatement and bring supporters to Nvidia’s headquarters. Nvidia’s chairman said the company had no plan to reinstate him. The reports documented Kalanick’s remarks but showed no strike or protest taking place.

    IT之家 AI · 🔥 0
  27. Nvidia Says AI Chip Deliveries and Product Road Map Remain on Track

    IT之家, citing Digitimes, reported that interim CEO Colette Kress said Nvidia's operations, business outlook and product roadmap have not been materially affected by Jensen Huang's departure. She said future development and launch schedules remain unchanged, market demand remains positive and the near-term disruption to AI-chip supply is limited. IDC said major customer commitments are already settled and cloud demand still exceeds supply, while warning that risks remain over the next two to four quarters.

    IT之家 AI · 🔥 0
  28. Report: Alibaba used offshore data centers to access Nvidia chips

    The New York Times reported that Alibaba used regional companies such as Megaspeed to access Nvidia chips through offshore data centers. It also reported that Alibaba provided AI computing technology to a laboratory described as having links to China’s military.

    Techmeme · 🔥 0
  29. DoNews reports Nvidia GPU rental prices rising, with B300 up over 60%

    DoNews reported that Nvidia GPU rental prices broadly increased, including a rise of more than 60% for the B300. The report also said China’s access to computing capacity was under pressure.

    DoNews · 🔥 0
  30. Nvidia Says Huang Will Not Return as Interim CEO Addresses Staff

    Interim CEO Colette Kress told Nvidia employees that the company “can handle” the current situation, following Jensen Huang’s departure after more than 30 years as chief executive. Nvidia said Huang, its co-founder and former CEO, will not return, while Kress reportedly promised December raises and additional equity benefits. The company also said its revenue and profit exceeded analyst expectations and raised its next-fiscal-year revenue outlook.

    cnBeta · 🔥 0

← PreviousNext →

Experience and discussion from the community

Share my NVIDIA experienceAsk about NVIDIA

Switchyard: An open-source library that helps an AI agent choose which model handles each request.

What problem it solves AI agents need to choose among models while balancing task success, cost, and latency for their workloads. Applications also need flexibility in where routing runs and who controls model clients, credentials, retries, and service operation. Key features Multiple routing algorithms — Routing can be configured with Auto, Task, Execution, or Composite strategies, including mode…

github.comhttps://github.com/NVIDIA-NeMo/Switchyard

vllm: A fast and easy-to-use library for LLM inference and serving.

What problem it solves Serving large language models requires efficient memory management, flexible model execution, and practical integration with existing models and APIs. vLLM addresses these needs with inference and serving capabilities spanning batching, caching, quantization, distributed execution, and multiple hardware platforms. Key features Efficient memory and serving — vLLM manages atte…

github.comhttps://github.com/vllm-project/vllm

skills: Portable instruction sets that teach AI agents how to use NVIDIA software for end-to-end workflows.

What problem it solves AI agents need product-specific instructions to use NVIDIA software reliably across physical AI, robotics, simulation, CUDA-X libraries, RAG, and platform workflows. The catalog centralizes these skills while keeping them maintained in their respective product repositories and synchronized through an automated pipeline. Key features Agent-discoverable skill catalog — The NVI…

github.comhttps://github.com/NVIDIA/skills

SkillEvaluator: An open-source tool that checks AI agent skills and measures their effect on task results.

What problem it solves AI agent skills need to be checked for quality, security, duplication, and their actual effect on agent behavior. Static checks alone cannot establish whether guidance improves task results, so evaluation must also support live comparison with and without a skill. Key features Three-tier evaluation pipeline — Independent tiers validate skill safety and structure, detect over…

github.comhttps://github.com/NVIDIA/SkillEvaluator

llama.cpp: Plain C/C++ implementation for local and cloud LLM and VLM inference.

What problem it solves Running large language and vision-language models can require substantial setup and may depend on the available hardware. The project aims to make inference accessible across a wide range of local and cloud hardware with minimal setup. Key features Minimal-setup inference — The project provides installation options for common platforms, including scripts, Docker, pre-built b…

github.comhttps://github.com/ggml-org/llama.cpp

TensorFold: A native language-model inference engine for serving checkpoints on Apple Silicon and NVIDIA GPUs.

What problem it solves Serving language models requires reading checkpoints, tokenizing requests, and running accelerated kernels while managing a local model server. TensorFold addresses these requirements with a native binary that can run on Apple Silicon and NVIDIA GPUs without requiring Python or MLX for serving. Key features Native Metal and CUDA inference — The Zig-based engine executes mode…

github.comhttps://github.com/ashhart/TensorFold

sglang-omni: A multi-stage serving runtime for omni, speech, and TTS models.

What problem it solves Interactive multimodal and speech systems often require generation to span heterogeneous stages with different compute patterns, dependency structures, and resource needs. Serving these pipelines requires coordinated stage lifecycles, inter-stage transport, model integrations, and a unified API surface. Key features Coordinated multi-stage runtime — Models generation as coor…

github.comhttps://github.com/sgl-project/sglang-omni

polars: An analytical query engine written in Rust for DataFrame and SQL queries.

What problem it solves Data analysis workflows often require expressive query composition, flexible execution models, and interoperability with columnar data. Polars addresses these needs by allowing the DataFrame API and SQL to share the same optimizer and engine, with support for large workloads and custom extensions. Key features Unified DataFrame and SQL Queries — Queries can be written with e…

github.comhttps://github.com/pola-rs/polars

FreeToken: An edge-native MoE serving engine for running frontier-scale open-weight models on personal and consumer hardware.

What problem it solves Running frontier-scale open-weight models locally requires an inference platform that can make efficient use of the heterogeneous hardware available on a personal device. The project addresses this by treating GPUs, CPUs, host memory, and interconnects as a unified, elastic inference platform. Key features Edge-native MoE runtime — The runtime provides MoE serving with bandw…

github.comhttps://github.com/FlashML-org/FreeToken

OpenShell: A safe, private runtime for fleets of autonomous AI agents.

What problem it solves Autonomous AI agents often need to read files, install packages, call APIs, and use credentials, but unrestricted access can expose host data, secrets, and networks. OpenShell provides these capabilities within an isolated runtime where declared policies constrain agent access. Key features Kernel-level runtime enforcement — Agents run in isolated sandboxes where kernel cont…

github.comhttps://github.com/NVIDIA/OpenShell

SkillSpector: Security scanner for AI agent skills.

What problem it solves AI agent skills execute with implicit trust and often receive minimal vetting before installation. SkillSpector helps users assess whether a skill contains vulnerabilities, malicious patterns, or other security risks before installing it. Key features Multi-format scanning — Scan agent skills supplied as Git repositories, URLs, zip files, directories, or individual files. So…

github.comhttps://github.com/NVIDIA/SkillSpector

jasna: Restores mosaics in videos and images through a simple app and command line that run on your GPU.

What problem it solves Mosaics obscure portions of videos and images, making it difficult to view the affected content without changing the rest of the picture. Users need a restoration workflow that can identify selected mosaic regions and process them with GPU-accelerated video handling. Key features Targeted mosaic restoration — Detects mosaics in videos and restores them while leaving the rest…

github.comhttps://github.com/Kruk2/jasna

Model-Optimizer: A library for composing model optimization techniques and exporting optimized quantized checkpoints.

What problem it solves Deep learning models often need to be adapted for efficient deployment across different inference frameworks. Model Optimizer addresses this by providing a unified set of optimization techniques and export workflows for quantized checkpoints. Key features Post-training quantization — Compress models with quantization techniques available for Hugging Face, Megatron-Bridge, Di…

github.comhttps://github.com/NVIDIA/Model-Optimizer

nvke: A self-hosting management plane for creating and operating Kubernetes clusters on NVIDIA GPU infrastructure.

What problem it solves Platform teams need to create and operate isolated Kubernetes clusters for customers on infrastructure they supply and control. They need a management plane that can provision GPU machines, manage cluster lifecycle operations, and establish node identity before issuing join credentials. Key features Independent workload clusters — Create workload clusters through the NVKE AP…

github.comhttps://github.com/dsx-ai-factory/nvke

LocalAI: An open-source AI engine for running models locally across modalities and hardware.

What problem it solves Running AI models locally can require integrating multiple inference engines, model formats, hardware accelerators, and API surfaces. LocalAI addresses this with a small core, on-demand backends, and compatible APIs that work across supported hardware without requiring a GPU. Key features Composable inference backends — Backends are packaged separately and pulled only when a…

github.comhttps://github.com/mudler/LocalAI

TensorRT-LLM: An open-source library for optimizing LLM and Visual Gen inference on NVIDIA GPUs.

What problem it solves LLM and Visual Gen inference requires specialized kernels, efficient runtime orchestration, and flexible deployment across NVIDIA GPU configurations. Developers also need a modular framework that supports customization while integrating with established inference tooling. Key features Optimized inference kernels — Custom kernels accelerate common inference operations, includ…

github.comhttps://github.com/NVIDIA/TensorRT-LLM

cuda-rust: NVIDIA's CUDA platform for Rust, providing host-side runtime crates and GPU kernel programming models in idiomatic Rust.

What problem it solves Rust developers need host-side access to the CUDA runtime and tools for expressing GPU kernels while preserving Rust's ownership and borrowing rules. The platform provides separate programming models for workloads that require direct thread-level control and workloads naturally expressed through tensor operations. Key features Shared CUDA runtime — Host-side crates provide d…

github.comhttps://github.com/NVIDIA/cuda-rust

auto-ontology: An open library that builds a context layer from enterprise metadata and exposes it to agent harnesses.

What problem it solves Agents can retrieve documents and query databases, but unreliable results undermine agentic analytics. Teams need a context layer that records where data lives, what its definitions mean, whether it is usable, how it should be queried, its provenance, and related concepts. Key features Enterprise ontology generation — Generate semantic ontologies from database metadata, infe…

github.comhttps://github.com/NVIDIA/auto-ontology