The Archive

Large language models (LLMs) are rapidly expanding their context windows, with recent models supporting sequences of 128K tokens, 256K tokens, and beyond.... Large language models (LLMs) are rapidly expanding their context windows, with recent models supporting sequences of 128K tokens, 256K tokens, and beyond. However, training these models with extended context lengths presents significant computational and communication challenges. As context lengths grow, the memory and communication overhead of attention mechanisms scale quadratically… Source

Enterprise AI strategy template and guidance on choosing automation approaches; generic business advisory content.

OpenAI outlines Sora feed design philosophy emphasizing personalized recommendations and parental controls.

In LLM training, Expert Parallel (EP) communication for hyperscale mixture-of-experts (MoE) models is challenging. EP communication is essentially all-to-all,... In LLM training, Expert Parallel (EP) communication for hyperscale mixture-of-experts (MoE) models is challenging. EP communication is essentially all-to-all, but due to its dynamics and sparseness (only topk experts per AI token instead of all experts), it’s challenging to implement and optimize. This post details an efficient MoE EP communication solution, Hybrid-EP, and its use in the… Source

Anthropic partners with Allen Institute and Howard Hughes Medical Institute to accelerate scientific discovery.

Import AI 443 examines agent ecology systems, Moltbook framework, and adversarial agent corruption risks.

OpenAI and Snowflake announce $200M partnership embedding frontier models and agents directly in Snowflake's data platform.

SpaceX acquires xAI, consolidating AI development with rocket/hardware infrastructure.

OpenAI releases Codex app for macOS enabling parallel multi-agent coding workflows with long-running task support.

NVIDIA CUDA Tile is a GPU-based programming model that targets portability for NVIDIA Tensor Cores, unlocking peak GPU performance. One of the great things... NVIDIA CUDA Tile is a GPU-based programming model that targets portability for NVIDIA Tensor Cores, unlocking peak GPU performance. One of the great things about CUDA Tile is that you can build your own DSL on top of it. This post shares the work NVIDIA is doing to integrate CUDA Tile as a backend for OpenAI Triton, an open source Python DSL designed to write DL kernels for GPUs. Source

Sparse tensors are vectors, matrices, and higher-dimensional generalizations with many zeros. They are crucial in various fields such as scientific computing,... Sparse tensors are vectors, matrices, and higher-dimensional generalizations with many zeros. They are crucial in various fields such as scientific computing, signal processing, and deep learning due to their efficiency in storage, computation, and power. Despite their benefits, handling sparse tensors manually or through existing libraries is often cumbersome, error-prone, nonportable… Source

AI coding agents enable developers to work faster by streamlining tasks and driving automated, test-driven development. However, they also introduce a... AI coding agents enable developers to work faster by streamlining tasks and driving automated, test-driven development. However, they also introduce a significant, often overlooked, attack surface by running tools from the command line with the same permissions and entitlements as the user, making them computer use agents, with all the risks those entail. The primary threat to these tools is… Source

Project Genie lets Google AI Ultra subscribers create and explore infinite interactive worlds via experimental prototype.

OpenAI describes internal data agent using GPT-5 and Codex with memory for reasoning over large datasets.

OpenAI announces retirement of GPT-4o, GPT-4.1, and o4-mini from ChatGPT effective February 13, 2026; API unaffected.

Taisei Corporation deploys ChatGPT Enterprise for internal HR and talent development workflows.

ServiceNow adopts Claude to power customer-facing apps and boost internal productivity.

NVIDIA Run:ai v2.24 introduces time-based fairshare, a new scheduling mode that brings fair-share scheduling with time awareness for over-quota resources to... NVIDIA Run:ai v2.24 introduces time-based fairshare, a new scheduling mode that brings fair-share scheduling with time awareness for over-quota resources to Kubernetes clusters. This capability, built on the open source KAI Scheduler that powers NVIDIA Run:ai, addresses a long-standing challenge in shared GPU infrastructure. Consider two teams with equal priority sharing a cluster. Source

This post introduces Dynamic Context Parallelism (Dynamic-CP), a scheduling approach in NVIDIA Megatron Core used for LLM post-training or DiT pre-training. It... This post introduces Dynamic Context Parallelism (Dynamic-CP), a scheduling approach in NVIDIA Megatron Core used for LLM post-training or DiT pre-training. It dynamically selects the CP size per microbatch to efficiently handle variable-length sequences, achieving up to 1.48x speedup on real-world datasets. In large-scale model training, an often-overlooked bottleneck arises from the… Source

Advances in AI architectures have unlocked multimodal functionality, enabling transformer models to process multiple forms of data in the same context. For... Advances in AI architectures have unlocked multimodal functionality, enabling transformer models to process multiple forms of data in the same context. For instance, vision language models (VLMs) can generate output from combined image and text input, enabling developers to build systems that interpret graphs, process camera feeds, or operate with traditionally human interfaces like desktop… Source

OpenAI launches €500K EMEA youth safety and wellbeing grant program for NGOs and researchers.

OpenAI publishes EU Economic Blueprint 2.0 highlighting AI adoption initiatives and partnerships across Europe.

Grok Imagine API offers video generation with stated advances in quality, cost, latency.

OpenAI details safeguards protecting agent data when clicking links, preventing URL exfiltration and prompt injection.

Recent advances in large-scale diffusion models have revolutionized generative AI across multiple domains, from image synthesis to audio generation, 3D asset... Recent advances in large-scale diffusion models have revolutionized generative AI across multiple domains, from image synthesis to audio generation, 3D asset creation, molecular design, and beyond. These models have demonstrated unprecedented capabilities in producing high-quality, diverse outputs across various conditional generation tasks. Despite these successes… Source