Leaderboard LLM

« Designed for intensive tasks, this model from Google’s Gemini family outperforms its predecessor in token efficiency while improving performance in coding and knowledge-based tasks. With a 1-million-token context, full multimodality, and better agentic task performance »

« The latest high-performance LLM model from xAI, specialized in reasoning and logic. This AI has achieved record results in academic and mathematical tests. Multi-agent version tackles the toughest challenges »

« An open-source model designed for code and long-running tasks, with a context window of up to 1 million tokens. This Mixed Expert (MoE) LLM handles complex projects, analyzes entire repositories, and performs multi-step reasoning. It is available as open source under the MIT license »

« Leverage powerful AI optimized for coding, finance, and autonomous tasks. This new LLM model from SpaceXAI is designed to be fast, cost-effective, and capable of handling complex end-to-end workflows. Trained with Cursor »

« Anthropic's latest high-end model, optimized for agentic code, long-running tasks, and critical business scenarios. It features self-checking of responses and a context window of one million tokens. It is also capable of automatically adjusting the depth of its reasoning based on the difficulty of the problem »

« This Mixture-of-Experts model, with approximately 1 trillion parameters, features a 1M-token context, virtually infinite Engram memory, and multimodal capabilities for text, images, and video. It aims to achieve performance scores comparable to Claude Opus while remaining significantly more affordable (to be released under the Apache 2.0 license) »

« Moonshot's LLM model, which rivals the best proprietary models on numerous benchmarks. Designed for long-form coding, advanced reasoning, and knowledge work tasks, it is also the first open-source model to surpass the 2.8 trillion-parameter mark »

« OpenAI's most powerful frontier model directly controls your computer, processes 1 million tokens, and reasons about very long tasks. It is claimed to reduce errors by 33% and significantly improve code and document analysis »