Full-text search across 66093 articles. Combine with topic and date filters; results sorted by relevance.
Anthropic released Claude Fable 5.1, which the company claims is its best model yet for coding, data analysis, and agentic work, with cache read prices cut from $1 to $0.25 per million tokens and reduced safety intervent…
LLMRix Inc. has released LLMRix Model Router, an open-source multi-model routing and orchestration framework for Java that manages provider differences, model selection, failover, quotas, costs, and observability. The fr…
Moonshot AI's open-weight Kimi K3 model matches frontier proprietary models on coding benchmarks for the first time, scoring in the range of GPT-4o and Claude Sonnet on SWE-bench Verified. The sparse mixture-of-experts m…
AMD has released an open-source Radeon GPU Detective (RGD) MCP Server that connects large language models to AMD's crash analysis pipeline, enabling automatic post-mortem GPU crash debugging. The tool allows developers t…
A new study from METR found that experienced developers using AI tools took 19 percent longer to complete tasks in their own large codebases, contradicting earlier Microsoft Research findings of a 55.8 percent speedup wi…
Anthropic's Claude Mythos Preview autonomously discovered over 23,000 software vulnerabilities via Project Glasswing, while the open-source community launched cost-optimization tools for AI coding agents and enterprise A…
Anthropic's Model Context Protocol (MCP) enables universal tool connectivity for AI agents, and Qwen3.6-35B-A3B, a Mixture of Experts model with 262k-token context, is optimized for MCP-based agentic tasks. A local GitHu…
Developer Yeo Kheng Meng created SwiftII, a Swift-flavored mini development environment for the Apple II series, bringing a subset of the modern Swift language to the 1977-era 1 MHz 6502-based computer. The project inclu…
AI and supply-chain governance are making bespoke software cheaper and safer than using third-party libraries, reversing the long-standing 'don't reinvent the wheel' advice. The cost of writing custom code has dropped du…
CLI authentication flows that rely on localhost redirects and browser availability fail in SSH sessions, containers, and WSL environments. The author argues that RFC 8252's OAuth pattern for native apps does not address …
Anthropic launched Claude Fable 5 and Claude Mythos 5, raising security and compliance concerns. A supply chain attack via compromised Microsoft GitHub infrastructure targeted AI coding agent users, bypassing traditional…
A developer has released tiny-vllm, a high-performance LLM inference engine written in C++ and CUDA that serves as a smaller sibling to the vLLM project. The open-source repository includes both the full source code for …
Non-technical "citizen builders" using generative AI tools to rapidly develop applications are creating systemic security exposures, as AI agents frequently recommend insecure configurations like public storage access an…
Lspkit is a Rust toolkit that serves both the Language Server Protocol used by code editors and the Model Context Protocol used by Claude Code, Cursor and Codex from a single engine, according to the project's blog post.…
Anthropic pushed a new usage rule to Claude Code around October 27, delivered from its servers under a flag named "tengu_heron_brook" with no user setting to disable it. The rule instructs the tool to "build original gam…
Small Python v0.4.0, released on GitHub by developer soasme, now runs at 0.98x–1.1x of CPython speed, up from 1.3x–1.5x slower, after a single Claude Code session with /remote-control enabled was used to benchmark and op…
A developer at Axonscape's Launchpad documented a common deploy failure in which builds break with "Could not resolve" errors for imports whose capitalization doesn't match the actual filename, even though the app runs f…
GitLab 19.5 has shipped with the GitLab MCP server reaching general availability for Claude Code and Cursor, alongside a Secrets Manager built on OpenBao and external agent governance controls. The release targets pipeli…
Paseo released a Vercel Sandbox plugin that lets users run the fx, Codex, Claude Code, OpenCode, and Pi coding agents in the cloud, starting a session from a laptop and controlling it from a phone even when the laptop is…
A developer released a tool for reviewing and understanding pull requests more easily, built to help users of AI-generated code keep focus on the parts of a PR that matter. The developer said they use AI to write softwar…