Latest in AI

Nordic data center operator atNorth on October 5, 2026 announced plans to develop FIN05, a new data center in Salo, Finland. The company expects to invest approximately €2 billion in the 28.6-hectare site, where power for a planned 75 MW first phase is secured, with a path to up to 230 MW for the wider development. Site Specifications and Investment Plans atNorth said Salo offers a strong location for critical infrastructure serving customers with demanding requirements for connectivity and…

One transformer ran candidate generation and ranking in Yandex Music's A/B test without hand-engineered features, lifting likes 11.42%.

The post Yandex Introduces Sona: A Single Generative Recommender That Replaces Entire Recommendation Cascade appeared first on MarkTechPost.

Alibaba's Qwen went from an invite-only chatbot in April 2023 to a 2.4-trillion-parameter open-weight model in August 2026. This is the full story, release by release: every major model, its key feature, and how its license changed. Each claim links to its source.

The post The Story of Qwen: Alibaba’s AI Models From 7B to 2.4T appeared first on MarkTechPost.

Cantina Security, with Yeta Labs, has released apex-flash-1, an open-weights model trained specifically for vulnerability research. It is a reinforcement learning fine-tune of Z.ai’s GLM-5.3-Flash, released on Hugging Face under the MIT license. Is it deployable? Yes, the MIT weights serve on vLLM, SGLang or Transformers, but BF16 needs roughly 640 GB of GPU memory. […]

The post Can an Open Model Do Security Research? Cantina’s apex-flash-1 Solves 40 of 60 Held-Out Bug Tasks appeared first on MarkTechPost.

Astra leads computer use, Argon leads legal and finance work, and Sol wins on price for coding agents.

The post GPT-6 Astra vs GPT-6.1 Sol vs Gemini 4 Argon vs Claude Fable 5.1: Which Frontier Model Fits Which Job appeared first on MarkTechPost.

President Donald Trump on October 4, 2026, announced the formation of the Super Intelligence Force in a post on his Truth Social account, naming four administration officials to lead a body he said will coordinate the federal government's effort to ensure America continues to lead the world in super intelligence. Mandate and Engagement Scope Trump presented the force as a follow-up to what the post calls the Historic White House Accord on Super Intelligence, an event at which, he wrote, the…

AI does everything at scale, from DOS-attack-level web scraping to producing, or enabling, such vast volumes of scientific research submissions that the result is becoming both a logistical crisis and a crisis of quality, as the signal-to-noise ratio continues to become unnavigable. Last week, preprint server arXiv announced that it will be instituting a new rate-limiting policy for submitters, who will now be capped at two submissions per month. The measure has been taken, the announcement…

AI guardrails are layered technical and procedural controls that constrain inputs, actions, outputs, and escalation around a model or agent. This guide explains the mechanism, trade-offs, evaluation, and controls that matter in practice.

Google Research has unveiled a federated learning system that moves gradient computation from phones into attested server-side TEEs. Access policies are published to Sigstore's Rekor log and the binaries are reproducibly buildable, so central differential privacy can be checked externally. Gboard already uses it for English and Japanese next-word prediction.

The post Google Research Moves Federated Learning Into TEEs: Gboard Now Trains With Externally Verifiable Differential Privacy appeared first on MarkTechPost.

Aleph Alpha has released Kolibri, a 78.1B-parameter English-German Mixture-of-Experts model that activates only 3.46B parameters per token. It has a 1M-token context and per-request reasoning effort, and its Apache 2.0 FP8 weights run on a single B200 or H200.

The post Aleph Alpha Releases Kolibri: A 78.1B Open-Weight English-German MoE Model With Only 3.46B Active Parameters appeared first on MarkTechPost.

DeepSeek has released official macOS and Windows desktop apps for DeepSeek Harness v0.2, its MIT-licensed agent harness. The preview adds a plugin manager, file and code-change review, and scheduled Automation Tasks. It also supports non-DeepSeek models through OpenAI-compatible endpoints.

The post DeepSeek Harness v0.2 Brings Official Desktop Apps to Its Open-Source Agent Harness appeared first on MarkTechPost.

Learn how NVIDIA IsaacTeleop turns XR hand tracking and motion controller input into robot commands using a pure Python retargeting engine and NumPy.

The post Inside NVIDIA’s IsaacTeleop: From Hand and Controller Tracking to Robot Actions with the Graph-Based Retargeting Engine appeared first on MarkTechPost.

Stochastic gradient descent and Adam are optimization algorithms that update model parameters from estimated gradients, but they use different rules for momentum and per-parameter step sizes. This guide explains the mechanism, trade-offs, evaluation, and controls that matter in practice.

TL;DR: Meta’s Muse, OpenAI’s Dots and Uber’s driver assistant share one bet: the agent speaks first. That moves the hard problem from what to answer to when to interrupt, on which channel, and with what offer. Classic ML and new decision models can solve it. Three launches, one pattern From pull to push Chatbots were […]

The post Meta, OpenAI and Uber Just Taught AI Agents to Talk First. What About When to Stay Quiet? appeared first on MarkTechPost.

IBM has made a self-hosted deployment of IBM Bob, its agentic software development platform, generally available. Enterprises can now run Bob on premises, in private or sovereign clouds, and in air-gapped networks. They bring their own model: NVIDIA Nemotron or Poolside Laguna for full isolation, or Claude, Gemini or GPT models through hybrid setups. Optional Premium Packages extend Bob to Java, IBM i and IBM Z modernization.

The post IBM Brings Bob to Self-Hosted and Air-Gapped Environments: Agentic Software Development Without Moving Your Code appeared first on MarkTechPost.

Prime Intellect has launched Prime Inference, an OpenAI-compatible platform for serving frontier open models on NVIDIA Blackwell. Its GLM-5.3 deployment uses Dynamo, vLLM and NVFP4 KV compression to serve 66 sessions per prefill group at 101 tok/s per user.

The post Prime Intellect Launches Prime Inference: Serverless and Reserved Serving for Frontier Open Models appeared first on MarkTechPost.

Microsoft AI has released MAI-Transcribe-2-Streaming, its first real-time speech-to-text model. It ranks #1 of 38 models on Artificial Analysis AA-WER Streaming. It scores 2.5% WER at 0.13s on final transcripts and 2.5% at 0.12s on first partials. It covers 60 languages with continuous language detection and costs $0.54 per hour during the introductory period. It is available now in public preview on Microsoft Foundry.

The post Microsoft AI Releases MAI-Transcribe-2-Streaming: #1 Real-Time Speech-to-Text Model on Artificial Analysis appeared first on MarkTechPost.

Decision AI models answer typed questions with calibrated probabilities instead of generated text. TypeSafe's Jev costs $0.042 per million input tokens and returns responses in 70 to 500 milliseconds. We break down how it works and its benchmarks, then compare it with Fastino's GLiDE, GLiNER2.5-Decide and 4 open-source rivals.

The post Decision AI Models Explained: TypeSafe Jev vs Fastino GLiDE, GLiNER2.5-Decide and Open-Source Competitors appeared first on MarkTechPost.

Decagon announced on October 2, 2026 that it is a launch partner for OpenAI Marketplace, OpenAI's new B2B marketplace, under which eligible OpenAI enterprise customers can apply part of their existing OpenAI commitment toward Decagon's customer-facing AI agents. The announcement was authored by Roger Liang, Decagon's strategic alliances and partnerships lead, and Ariana Xiang, a senior product marketing manager at the company. Decagon said the partnership makes it easier for businesses building…

Meta on October 2, 2026 open-sourced the firmware and device SDKs for building Muse gadgets, do-it-yourself hardware that connects the company's Muse AI agent to displays, buttons, sensors, and actuators, and introduced the Muse Home Link, a USB-C device that links Muse to a home network, offered free to U.S. Muse subscribers while supplies last. Nat Friedman announced the release in a post on X, writing, "Announcing Muse Gadgets, an open source ESP32 firmware and Linux sdk so that you can make…