Trending blocks

Gemini's newest multimodal model, with next generation features and improved capabilities

Hybrid reasoning model with superior intelligence for high-volume use cases and a 200K context window

The latest series of Code-Specific Qwen models, with significant improvements in code generation, code reasoning, and code fixing

Qwen3-Coder-480B-A35B-Instruct is a Mixture-of-Experts (MoE) code generation model developed by the Qwen team. It is optimized for agentic coding tasks such as function calling, tool use, and long-context reasoning over repositories. The model features 480 billion total parameters, with 35 billion active per forward pass (8 out of 160 experts).

The smartest, fastest, most useful OpenAI model yet, with built-in thinking that puts expert-level intelligence in everyone’s hands.

An upgrade to Claude Opus 4 on agentic tasks, real-world coding, and reasoning

OpenAI's versatile, high-intelligence flagship model

Knowledge graph-based persistent memory system

Anthropic's most intelligent model

Local filesystem access with configurable allowed paths

This server enables LLMs to interact with web pages, take screenshots, and execute JavaScript in a real browser environment using Playwright.

References the currently open file, providing quick access to its contents for immediate review or modification.

References the contents of all your open files, providing a quick overview of the active code across your development environment.

With Browser MCP, you can use MCP to automate your browser so that AI applications can navigate the web, fill out forms, and more. https://browsermcp.io/

References the contents from any documentation site, providing easy access to relevant information for quick guidance and development support.

GitHub repository access and management

Reference problems in the currently open file

DeepSeek-R1 achieves performance comparable to OpenAI-o1 across math, code, and reasoning tasks

Shows the structure of your codebase, making it easier to navigate files and directories.

DeepSeek-R1 achieves performance comparable to OpenAI-o1 across math, code, and reasoning tasks

References specific functions or classes from throughout your project, providing quick access to key code elements for review or modification.

Kimi K2 Instruct is a large-scale Mixture-of-Experts (MoE) language model developed by Moonshot AI, featuring 1 trillion total parameters with 32 billion active per forward pass. It is optimized for agentic capabilities, including advanced tool use, reasoning, and code synthesis. Kimi K2 excels across a broad range of benchmarks, particularly in coding (LiveCodeBench, SWE-bench), reasoning (ZebraLogic, GPQA), and tool-use (Tau2, AceBench) tasks.

Web and local search using Brave's Search API

OpenAI's fast, affordable small model for focused tasks

References the operating system context, providing details about the current OS environment for troubleshooting or configuration purposes.

The Exa MCP server enables AI assistants to perform web searches using Exa's powerful search API, providing structured search results including titles, URLs, and content snippets

References the commit history of your current branch, providing an overview of all past commits for tracking changes and project progress.

A high-performing open embedding model with a large token context window

For Grok 4, we utilized Colossus, our 200,000 GPU cluster, to run reinforcement learning training that refines Grok's reasoning abilities at pretraining scale