Author: IBL News

IBL News | New York

Nvidia has agreed to buy Hugging Face for $12.9 billion, The Information reported. The companies avoided commenting on the transaction.

Hugging Face, founded in 2016, is one of the most popular hubs where developers share and download open-source AI models.

Buying it will give Nvidia a strong presence in the open-source AI space, helping to protect its dominance in AI chips, especially since the biggest closed-source companies (OpenAI, Google, Amazon, and Anthropic) are now building their own AI chips to reduce their reliance on Nvidia.

Also, owning Hugging Face — which already helps developers run their AI models using rented computing power — could give Nvidia a way back into the cloud computing market without starting from scratch, since Nvidia reportedly scaled back its own cloud business, called DGX Cloud, about a year ago.

On its side, Nvidia has already poured tens of billions of dollars into building its own open-source AI models.

Not surprisingly, Hugging Face CEO Clem Delangue has spent much of this year publicly aligned with Nvidia’s open-source push. In this regard, Hugging Face used an Nvidia-modified version of a Chinese open-source model to defend itself after a cyberattack.

On the other hand, Chinese labs like Moonshot AI had released systems — such as its Kimi K3 model — that matched leading U.S. models on benchmarks while costing much less to run.

The price marks a huge jump from Hugging Face’s last known value. The company raised $235 million in a 2023 funding round, valuing it at $4.5 billion. That round was led by Salesforce Ventures, with money also coming from Alphabet’s GV, IBM Ventures, and Nvidia, among others.

Hugging Face generates only about $150 million in annual revenue, up from roughly $100 million just two months earlier.

Nvidia acquires Hugging Face for ~$12.9bn. But who gets the 💰? My usual breakdown below 👇

*Investors are sharing ~$7.3bn of profits on <$400m invested. A cool ~20x blended.*

*1) The single biggest winner here? Lux Capital.* 👑

Lux is taking the crown for the most $ returned… pic.twitter.com/AScos7vSee

— Paul (@P_Bonnet) August 27, 2026

IBL News | New York

Perplexity announced a version of its agentic “Computer platform”, named Portable Computer, that runs on hardware owned by clients, such as NVIDIA’s DGX Spark and Linux-powered systems with RTX GPUs.

The launch, developed in close partnership with Nvidia, reflects a new trend of AI agent workloads moving off the cloud and onto local devices with no billing credit consumption.

This local app has the same UI as Perplexity’s cloud version.

Portable Computer will be offered for Pro, Max, Enterprise Pro, and Enterprise Max subscribers on Linux, with Windows support following in September. Any RTX GPU with at least 24GB of VRAM — roughly a GeForce RTX 3090 or newer — clears the bar.

While Perplexity Computer orchestrates AI models, files, tools, and web access to complete multi-step tasks — reviewing folders of documents, analyzing data, producing reports, and pushing results into business systems. Portable Computer replicates that experience locally: the local models, agent harness, inference engine, tools, app connectors, and a security sandbox come packaged together in a single system.

Typically, today, with local AI stacks users must assemble and operate those pieces separately — downloading model weights, standing up an inference server, wiring together tools, and tuning performance.

Venture Beat did a demo running a 27-billion-parameter Qwen model at full GPU utilization on a DGX Spark, the agent reviewed each document and flagged cases where the hypothetical investor was paying unnecessary fees.

Alongside the launch, Perplexity published a research paper arguing that effective local agents require the model and the agent harness — the scaffolding of prompts, tools, and orchestration logic around the model — to be designed together. The core insight: general-purpose harnesses assume a frontier model that can absorb enormous contexts, navigate sprawling tool surfaces, and plan over long horizons. Small local models buckle under those demands.

Perplexity found empirically that although models like Qwen 3.8 27B advertise 260,000-token context windows, they begin to struggle beyond 100,000 tokens. So the company built a deliberately minimal harness: a succinct system prompt, a small set of core tools, and capabilities that load and unload as on-demand “skills” rather than sitting permanently in context.