b9781
vulkan: allow reducing the graph submission batches to avoid timeouts (#24872) macOS/iOS: macOS Apple Silicon (arm64) macOS Apple Silicon (arm64, KleidiAI enabled) DISABLED macOS Intel (x64)…
Latest
Databricks positioned highest in execution and furthest in vision for the second consecutive year in Gartner Magic Quadrant
Enterprises are rapidly deploying agentic applications at scale, from back-office...
New in Runway, you can now localize ads. One image in, any language out. Input a single ad and get a version for every market. All with a single click…
New in Runway, you can now localize ads.One image in, any language out. Input a single ad and get a version for every market. All with…
OpenAI unveils its first custom chip, built by Broadcom
Named Jalapeño, the new processor was designed specifically for the unique needs of OpenAI's inference systems.
For Most of the World, Open-Source AI Is the Only Way Forward
Article URL: https://techstrong.ai/articles/for-most-of-the-world-open-source-ai-is-the-only-way-forward/ Comments URL: https://news.ycombinator.com/item?id=48660839 Points: 12 # Comments: 0
biorisk *can* be solved. It's harder than cyber but doable well, actually not. There is a vastly less tractable attack angle of bioterrorism against n…
biorisk *can* be solved. It's harder than cyber but doablewell, actually not. There is a vastly less tractable attack angle of bioterrorism against non-human life. But…
RubyLLM: A single, beautiful Ruby framework for all major AI providers
Article URL: https://rubyllm.com/ Comments URL: https://news.ycombinator.com/item?id=48660711 Points: 4 # Comments: 0
Running Windows Games on a Hobby OS with Wine
Article URL: https://astral-os.org/posts/2026/04/03/wine-on-astral.html Comments URL: https://news.ycombinator.com/item?id=48660671 Points: 9 # Comments: 2
OpenAI reveals its first AI processor: Jalapeño
OpenAI has just revealed a new "intelligence processor" chip for AI servers made in partnership with Broadcom. The chip, called Jalapeño, is designed to power current…
PR spam today looks like email spam in the early 2000s
Article URL: https://www.greptile.com/blog/prs-on-openclaw Comments URL: https://news.ycombinator.com/item?id=48660579 Points: 21 # Comments: 9
My micro-benchmark: how good are LLMs at simulating wetting behaviour?
Example surfaces that LLMs are asked to simulate, showing simulated liquid (green) shaped by solid constraints (orange). Overall score, pass count, and recorded token/cost totals for…
…is Elon in cahoots with the CCP?
…is Elon in cahoots with the CCP?Truth Seeker: X is actively suspending accounts that are posting Chinese caste content.
Semantic Browsing: Controllable Diversity for Image Generation
Modern text-to-image models excel in visual fidelity and prompt adherence. However, this strict adherence comes at the cost of diversity: generated samples tend to collapse into…
AGORA: An Archive-Grounded Benchmark for Agentic Workplace Document Reasoning
Large language models are increasingly deployed as agents that reason over documents rather than answer from parametric knowledge. We study archive-grounded reasoning: locating sparse evidence across…
OpenAI and Broadcom unveil LLM-optimized inference chip
https://openai.com/index/openai-broadcom-jalapeno-inference-chip/ Quoted from the start of the blog post: Early testing shows that the first-generation accelerator will deliver performance per watt substantially better than current state-of-the-art…
ChartWalker: Benchmarking the Cross-Chart RAG Task
Cross-Chart Retrieval-Augmented Generation (RAG) is critical for complex multi-modal analytical tasks in scientific, business, and political domains. However, existing benchmarks either focus on tables, which are…
QG-MIL: A Gated Transformer Aggregator for Domain-Agnostic Multiple Instance Learning in Medical Imaging
Attention-based Multiple Instance Learning aggregators in medical imaging are prone to attention concentration, producing overconfident and unstable predictions. We introduce QG-MIL, a gated transformer aggregator that…
EventVLA: Event-Driven Visual Evidence Memory for Long-Horizon Vision-Language-Action Policies
Memory remains a critical bottleneck for long-horizon robotic manipulation, as standard Vision-Language-Action (VLA) policies often fail when task-relevant cues become occluded or unobservable over time. While…
The Swiss Federal Supreme Court is evaluating Heretic
“Oh no, are they banning abliterated models now?!?” If that was your first thought when you read the title I can’t blame you. But that’s actually…
LingxiDiagBench: A Multi-Agent Framework for Benchmarking LLMs in Chinese Psychiatric Consultation and Diagnosis
Mental disorders are highly prevalent worldwide, but the shortage of psychiatrists and the inherent subjectivity of interview-based diagnosis create substantial barriers to timely and consistent mental-health…
FLAT: Feedforward Latent Triangle Splatting for Geometrically Accurate Scene Generation
Generating explorable 3D scenes from a single image requires strong generative priors and accurate geometric representations suitable for downstream use. Current video diffusion models offer high-quality…
AOHP: An Open-Source OS-Level Agent Harness for Personalized, Efficient and Secure Interaction
AI agents are driving a new software paradigm, with the ability to autonomously call tools, extract information, manage memory, and complete tasks that span applications and…
Escaping the Self-Confirmation Trap: An Execute-Distill-Verify Paradigm for Agentic Experience Learning
Experience-driven self-evolution is critical for large language model (LLM) agents to improve through open-world interaction. However, existing experience learning methods mostly rely on single-agent loops, where…
FlowR2A: Learning Reward-to-Action Distribution for Multimodal Driving Planning
Multimodal driving planning faces a long-standing tension between two paradigms: scoring-based methods benefit from dense reward supervision but are confined to a fixed action vocabulary, while…