Search “NVIDIA”

Schneider Electric, Nvidia and HPE Partner to Launch Industrial AI Agents

Schneider Electric, NVIDIA and HPE have established a strategic partnership to launch the first Industrial AI Agents at the Edge for industrial operations, focusing on operational technology (OT) monitoring, cybersecurity and energy consumption optimization. These agents run locally on HPE Private Cloud AI, eliminating the need to continuously transmit sensitive industrial data to the public cloud; customers can manage the agents through Schneider Electric's EcoStruxure Industrial Insights application. The related OT monitoring technology has been in use for three years at Chilean pulp company CMPC. There are currently no subsequent developments.

View story:Schneider Electric, NVIDIA and HPE Partner to Launch Industrial AI Agents

Perplexity develops post-training method JOLT

Perplexity unveils post-training method JOLT, letting the same model serve as both teacher and student, with the teacher also participating in training so that feedback is always available. Perplexity's release states that on MATH, JOLT reaches accuracy comparable to GRPO with 6.5 times fewer training tokens. A subsequent MarkTechPost report states that the method was proposed by Perplexity and NVIDIA, named JOLT (Joint On-policy Learning and Teaching); JOLT+ achieves a 72.09% state-test pass rate on AppWorld, 9.9 percentage points higher than GRPO's 62.20%. The report states that testing covered Qwen3-0.6B, Qwen3-4B-Instruct, Qwen3-8B, Qwen3-30B-A3B-Instruct and DeepSeek-V2-Lite, and that the code has not yet been released.

Hao AI Lab announces that FastH3 Omni can run locally on DGX Spark, RTX GPUs, and Apple MLX

Hao AI Lab announced that FastVideo FastH3 Omni can run locally on NVIDIA DGX Spark, RTX GPUs, and Apple MLX. The lab's reported performance figures include: generating a 5-second 480p video from a single photo and a prompt takes about 3 minutes, and about 2 minutes on two Sparks; FastH3 T2AV takes 52.6 seconds on a single Spark and 30.7 seconds on two Sparks. The clips in the demo video were all generated by DGX Spark using an 8-step NVFP4 checkpoint. The lab also provided links to a blog, a Cookbook, and Hugging Face. There are currently no subsequent developments.

Apoorv Chaudhary joins discussion on local AI and privacy

On October 22, 2026, episode 14 of OpenClaw's The ClawCast invited NVIDIA's Apoorv Chaudhary to join hrudolph and Pat_Erichsen in discussing local AI, privacy, hardware limitations, and the scenarios where cloud models still apply. The episode was released by OpenClaw and is currently the only confirmable source of facts. There are no subsequent developments at present, and the status of the event remains this podcast discussion itself.

Flapping Airplanes Completes $1 Billion Financing

Flapping Airplanes Inc. announced the completion of $1 billion in new financing. According to a SiliconANGLE report on October 22, 2026, Bloomberg said its valuation reached $6 billion; the company was founded about a year ago and has no commercial product yet. The report said that in its first year it mainly invested in research and development of new GPU primitives, and its megakernel framework can simultaneously complete full training, inference, and game running within a single kernel, allowing memory-bound inference and compute-bound training to overlap on the same chip. The Series A round was co-led by Index Ventures and Lux Capital, with Sequoia Capital and GV, investors in the $180 million seed round in January, continuing to participate, and Nvidia and others also on the list of institutional investors. There are currently no subsequent development facts.

Era4 executive proposes achieving financial sovereignty through domestic AI infrastructure

On October 22, 2026, Era4 executive Karl Havard put forward the proposition of achieving financial sovereignty through UK domestic AI infrastructure. According to related reports, Era4 advocates using self-built energy, UK domestic AI facilities, and proprietary hardware to give the UK more control over the AI technology stack and to keep related revenue within the domestic economy. The reports also mention that its Chesterfield facility uses liquid-cooled Nvidia B300 systems supplied by HPE and runs HPE Private Cloud AI; the company also plans to use about 20% of a 230-acre former steelworks site in Newport, Wales for AI infrastructure. The proposition is currently still at the proposal stage, with no subsequent developments of fact.

Inception releases diffusion language model Mercury 3

On October 21, 2026, Inception released the diffusion language model Mercury 3. According to a report by The Decoder the same day, Inception claims the model won 8 of 12 benchmarks and runs four to eight times faster than GPT-6 Luna and Claude Haiku 5.5. In Inception's own testing, Mercury 3 scored 76.84 on IFBench, 15 points higher than GPT-6 Luna; it scored 51.08 on SWE-Bench Atlas, while its competitors scored 33.33 and 32.26 respectively. Claude Haiku 5.5 led on the DeepSWE long-code test and long-text retrieval. The model generates 1,068 tokens per second on Nvidia GPUs, is priced at $0.20 per million input tokens and $0.70 per million output tokens, is available at 50% off during the launch period, and is now live on the Inception platform as well as OpenRouter, Baseten, and Vercel. The report notes that there are currently no independent tests.

Databricks AI Runtime is now generally available

On October 21, 2026, Databricks announced the general availability of Databricks AI Runtime (AIR). According to information published on the Databricks Blog the same day, developers can train and fine-tune open models on serverless NVIDIA GPUs without having to manage GPU infrastructure themselves. AIR supports NVIDIA NeMo, NGC containers, and Nemotron cookbooks, as well as multi-node distributed training, with the underlying clusters and networking managed by Databricks. Trained models can be deployed for real-time inference through Databricks Model Serving, and access to training data is governed by Unity Catalog. This news currently comes from a single source, the official Databricks blog, and there are no other independent reports or subsequent developments yet.

Daytona Completes $60 Million Series B Funding Round

Daytona, which provides cloud infrastructure for AI agents, has completed a $60 million Series B funding round. According to a Tech.eu report on October 20, 2026, the round was led by FirstMark, valuing the company at $1 billion, with cumulative funding reaching $91 million. Its platform provides isolated Linux, Windows, and macOS computing environments for agents through a single API, supporting task pausing, resuming, and continued exploration, with customers including Anthropic, OpenAI, NVIDIA, Cursor, and Cognition. The company says its revenue run rate grew 25-fold from January to September, and the new funds will be used to expand its agent cloud platform, scale up computing capacity, and hire in San Francisco, Croatia, and Singapore. There are currently no subsequent developments.

DeepSeek open-sources inference framework DeepInfer

On October 20, 2026, DeepSeek announced the open-sourcing of its internal production-grade inference framework, DeepInfer. According to Latent Space, the framework powers DeepSeek-V4.1-Flash, is customized based on vLLM, supports distributed inference, speculative decoding, SWA Bounded Replay, and structured output, and targets the NVIDIA SM100 and Huawei Ascend 950 platforms; the published single-cluster production figures are 155 trillion tokens processed per day, a 96% input cache hit rate, 400 ms first-token latency, and a generation speed of 230 tokens/s. The report also noted that components such as KV cache management and load balancing have not yet been fully open-sourced. The next day, DeepSeek published the DeepInfer technical report on Zhihu, disclosing internal details including the hardware selection for Prefill and Decode and the KVCache targets, and providing data showing a single cluster serving 155 trillion Tokens/day at $38 per card per hour at off-peak prices. The framework is currently open-sourced, but the KV cache system, Prefill/Decode disaggregation, and load balancing required for a complete service are not fully included in this open-source project.

Shopify shares how Sidekick uses a post-trained open-source model as the GraphQL sub-agent

On October 21, 2026, the Shopify engineering team, in a talk at NVIDIA GTC Berlin 2026, shared the model division-of-labor plan for its e-commerce assistant Sidekick: a frontier model handles conversation planning, while a self-deployed open-source model handles the GraphQL sub-agent. The team chose the MoE model Nemotron 3.5 Lightning, with 30 billion total parameters and 3 billion parameters activated per token; after 6 rounds of post-training and 61,000 repair samples, the review score rose from 37.7 out of the box to 74.5, slightly above the frontier model's 73. On GraphQL tasks, the median response was 0.87 seconds, versus 6.50 seconds for the frontier model API, at roughly 90% lower cost. Currently this talk is a single-source report, with no follow-up developments yet.

Porsche and others announce joint construction of a shared data flywheel for autonomous driving

Porsche, Volkswagen Group brands and CARIAD announce the joint construction of a shared data flywheel for autonomous driving development. The news was announced on October 21, 2026 at the NVIDIA GTC Berlin conference, with Porsche leading initial implementation and ecosystem integration, and Voxel51, NVIDIA and Microsoft participating. According to Voxel51, the group's connected fleet generates data each day equivalent to 11,000 ordinary drivers, and the goal of the cooperation is to reduce repetitive real-vehicle testing by up to 90% and shorten AI regression testing cycles by up to 85%. In terms of division of labor, Voxel51 provides the data layer and supports six workflows including multimodal retrieval, automatic annotation and scenario filtering, NVIDIA provides tools such as Cosmos Data Factory and Omniverse NuRec, and Microsoft Azure provides cloud and computing power. The cooperation is currently at the announcement stage, with no subsequent progress disclosed.

Mercedes announces point-to-point automated driving in production cars from 2027

On October 21, 2026, Mercedes announced that from 2027 it will offer automated driving from start to destination in production cars, with the driver still required to monitor at all times and bear responsibility. According to heise online, the system is equipped with 10 cameras, 5 radars and 12 ultrasonic sensors, a computing platform with up to 254 TOPS, and combines MB.OS with Nvidia Hyperion, Halos and the Alpamayo Stack. The report also states that at the end of 2026 it will first launch for selected customers in Stuttgart and Munich, that existing owners with hardware support can upgrade via OTA, and that a one-month free trial will be offered. The plan is currently at the announcement stage, and there is no qualified factual evidence of subsequent implementation progress.

Keysight executive argues AI data centers require system-level validation

On October 21, 2026, Keysight Korea President Lee Sun-woo argued that AI data centers need end-to-end system-level validation to ensure profitability. Citing Nvidia Rubin GPUs as an example, he said that if HBM data supply is insufficient, GPUs will sit idle rather than deliver compute. The solutions proposed by Keysight span chip design to data center networking, including the W3650B, W9100E EDA tools, XR8 oscilloscope, FITS, AresONE Plus 1600GE and KAI-IB, among which the XA6 uses GPUs to speed up wideband signal measurement by up to 50 times. This argument was reported by The Elec; there are no subsequent developments as facts.

English translation is being prepared.

English translation is being prepared.

English translation is being prepared.

English translation is being prepared.

Vultr announces deployment of a 20MW AMD MI355X GPU cluster in Chennai

Cloud infrastructure company Vultr announced that it will deploy a 20MW AMD MI355X GPU cluster in Chennai, India. According to an Analytics India Magazine report on October 21, 2026, this will be India's largest AMD Instinct GPU deployment and one of the largest in the world. The report said the MI355X was released last June, features 288 GB of HBM3E memory, higher than the Nvidia B200's 180 GB. Vultr currently operates 34 cloud data center regions in 17 countries worldwide and plans to further expand next year, introducing next-generation MI455X GPUs and the Helios rack-scale AI architecture. The deployment plan is currently in an announced state, with no qualified factual evidence of subsequent progress.

SGLang natively supports and powers the Beam production API

On October 16, 2026, LMSYS announced that SGLang now natively supports Reflection's Beam model and is being used to power its production API. Beam is a 501B-parameter MoE model with 23B active parameters per token and a native 1M-token context window, planned for release under Apache 2.0; SGLang provides BF16, NVFP4, and dynamic FP8 support for it, along with DFlash2 speculative decoding and hybrid attention prefix caching. Running OpenHands multi-turn agentic coding replay with NVFP4 on 4 NVIDIA GB300 GPUs, it reached 127.6K tokens/sec per GPU and 185.87 tokens/sec per user (p50) at a concurrency of 32. On the same day, Reflection released the Beam model weights, with Ollama offering them free for a limited time as a day-zero partner with zero data retention; Beam has approximately 501 billion total parameters, about 23 billion active parameters per token, a 1 million-token context window, and is released under the Apache 2.0 license. Subsequent information from AWS states that the Beam weights have been released on Hugging Face, offering three checkpoints—BF16, FP8, and NVFP4—and are now available on Amazon SageMaker JumpStart, and can also be accessed via SageMa