Baidu released Unlimited OCR, an open-source 3-billion-parameter model that uses Reference Sliding Window Attention to parse multi-page documents in a single pass, eliminating the need for page-by-pag…
The launch of Z.ai's GLM 5.2 and DeepSeek V4 Flash has created a 50x price gap between open-weight APIs and closed frontier models, reshaping the build-versus-buy calculus for developers. While open-w…
Anthropic accused Alibaba of executing the largest known distillation attack against its Claude AI model, involving 28.8 million queries from 25,000 fraudulent accounts between April and June 2025. Th…
NVIDIA's Rubin architecture and DSX AI factory reference design introduce a 100% liquid-cooled system operating with coolant temperatures up to 45°C, eliminating data center water consumption and redu…
PostHog engineer Robbie Coomber rebuilt the company's SQL parser using AI agents, replacing an ANTLR-generated parser with a hand-rolled recursive-descent parser that runs 70 times faster. The project…
NVIDIA released NeMo AutoModel, a library that integrates Expert Parallelism and DeepEP into Hugging Face's API, achieving 3.4x to 3.7x higher training throughput and 29% to 32% lower GPU memory consu…
CAPTCHAs are no longer effective at stopping bots, as modern AI solves them faster and more accurately than humans, forcing developers to adopt passive, cryptographic, and behavioral security measures…
OpenAI has taped out its first custom inference chip, named Jalapeño, built with Broadcom. The ASIC is designed to reduce inference costs for serving LLMs, following the vertical-integration strategy …
Yann LeCun warned at the United Nations Open Source Week that proprietary AI systems controlled by a few tech giants threaten linguistic diversity, cultural representation, and democracy. Open-weight …
A peer-reviewed critique in Nature reveals that Microsoft's 2025 quantum breakthrough, claiming observation of Majorana particles, was based on Python programming errors. Dr. Henry Legg of the Univers…
RubyLLM 1.16, a unified AI framework for Ruby, has been released with concurrent tool execution and Rails-native instrumentation, addressing performance and observability needs for enterprise applicat…
OpenAI and Broadcom co-developed Jalapeño, a custom AI inference chip designed to replace general-purpose GPUs for large language model serving. The chip, built in nine months, targets memory and netw…
Haystack, an open-source AI orchestration framework by deepset, uses explicit directed acyclic graphs (DAGs) instead of implicit chains to build production-ready RAG systems. Its modular, serializable…
Alibaba's Qwen team released Qwen-AgentWorld, a language model that simulates complex environments natively, replacing external simulators for training AI agents. The model, trained on over 10 million…
The privileged status of vulnerability reports has eroded as cheap LLM-generated bug reports flood security inboxes, shifting the bottleneck from finding issues to triaging noise. Maintainers now face…
Google fired a 7-year veteran engineer, Justin Poehnelt, after he built an open-source, agent-ready CLI for Google Workspace that went viral on Hacker News. The tool, now hosted under Google's own Git…
Japan's Sakana AI launched Fugu, a multi-agent system that coordinates a pool of frontier LLMs behind a single OpenAI-compatible API, achieving top benchmark scores on SWE-Bench Pro, LiveCodeBench, an…
A coalition including Cloudflare, Mozilla, Google, Microsoft, and Shopify has proposed PACT (Private Access Control Tokens) to replace CAPTCHAs and verify human web traffic without tracking users or r…
Cloudflare, major browser vendors including Mozilla Firefox, Google Chrome, and Microsoft Edge, and Shopify are developing the Private Access Control Tokens (PACT) protocol to replace invasive bot det…
Developers can deploy Llama 3.1 8B behind vLLM's OpenAI-compatible API on a rented GPU instance, achieving thousands of output tokens per second through continuous batching. The tutorial covers instal…