OpenAI ships GPT-5.1 with faster tool-calling
The update focuses on latency and tool-use reliability rather than raw benchmark gains, with early adopters reporting fewer malformed function calls.
Read the storyLatest Stories
Anthropic doubles context window for Claude API
Enterprise customers get a 400K-token context option, aimed at whole-codebase and long-document workflows.
We tried four vector databases so you don't have to
A practical comparison of Pinecone, Weaviate, Qdrant, and pgvector for a mid-size RAG pipeline, with real latency numbers.
Google expands AI Overviews to shopping queries
Retailers are watching click-through rates closely as AI-generated summaries move into commercial search results.
A practical guide to RAG that actually works in production
Beyond the tutorial: chunking strategy, retrieval evaluation, and the failure modes that only show up at scale.
EU AI Act's next compliance deadline lands in October
High-risk system providers have until Q4 to complete conformity assessments — here's what's actually enforced first.
Meta open-sources a smaller Llama variant for edge devices
The 3B-parameter model targets on-device inference, trading capability for a footprint that runs on phones.
Prompt caching cut our LLM bill by 60%
How restructuring prompts to put static content first let us take advantage of provider-side caching with almost no code change.
Amazon adds agentic checkout to Alexa+
The assistant can now complete multi-step purchases end-to-end, raising fresh questions about accidental orders.
Building your first MCP server: a complete walkthrough
Model Context Protocol lets any tool talk to any AI assistant. Here's how to wrap an internal API as an MCP server in under an hour.