Buzz & Trends

Frontier general-purpose models with strong reasoning and tool use.

Long-context models known for coding, writing, and safety.

Multimodal models tightly integrated with Google's ecosystem.

Open-weight models you can self-host and fine-tune.

Efficient open and commercial models from Europe.

Strong open reasoning models at low cost.

Capable open-weight multilingual model family.

Open standard that lets AI apps connect to tools and data sources.

Reusable connectors (GitHub, Postgres, filesystem, web) for any MCP client.

Build stateful, multi-step agent workflows as graphs.

Framework for orchestrating role-playing autonomous agents.

Multi-agent conversation framework for complex tasks.

In-editor AI pair programmer and chat across IDEs.

AI-first code editor with whole-repo context and agents.

Agentic coding in the terminal that edits and runs your repo.

Agentic IDE with flows that plan and edit across files.

Generate React/Tailwind UIs from prompts.

Prompt a full-stack app in the browser and deploy it.

Toolkit for chaining LLM calls, tools, and retrieval.

Data framework for RAG and knowledge-augmented LLMs.

TypeScript SDK for streaming AI UIs and tool calls.

Run open LLMs locally with one command.

Hub for models, datasets, and Spaces the GitHub of ML.

Managed vector database for semantic search and RAG.

Open-source vector search engine you can self-host.

Lightweight open-source embedding database.

Vector search inside Postgres no extra service.

Frontend cloud for Next.js with instant global deploys.

Edge compute, Workers AI, R2 storage, and CDN.

The broadest cloud compute, storage, AI (Bedrock), and more.

Deploy apps and databases with zero config.

Run app servers close to users, globally.

Package apps into portable containers.

Orchestrate containers at scale.

Infrastructure as code across any cloud.

CI/CD pipelines that live next to your code.

Dashboards and observability for metrics and logs.

Open-source Firebase: Postgres, auth, storage, realtime.

Serverless Postgres with branching.

Drop-in auth and user management.

Developer-first email API.

Serverless Redis and Kafka, pay-per-request.

One API to call dozens of LLMs, with free models.

Ultra-fast LLM inference with a free tier.

High-quality AI image generation.

Realistic AI voice and text-to-speech.

Open speech-to-text that runs anywhere.