Codestral

« Take advantage of Anthropic’s AI model—more reliable, more honest, and four times less likely to overlook its own errors. Its fast mode delivers up to 2.5 times more tokens per second (at the same API rate as the previous version) »

« Choose between two models optimized for professional work, coding, and using tools. They improve speed and significantly reduce costs compared to the previous generation »

« A frontier model from Meta built for personal and professional use, featuring an active memory that remembers what matters to you. It can navigate independently between multiple apps to see your tasks through to completion. Includes robust coding capabilities as well »

« 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 »

« A fast and cost-effective model designed for code, agents, and complex workflows, with a 1-million-token context and significant improvements on long-running tasks and multi-step execution. Compared to its predecessor, this model scores 9 points higher on FrontierCode 1.1 and 16 points higher on DeepSWE v1.1 »

« A reasoning model that is twice as fast and half the cost, works longer on difficult tasks, checks its own work more thoroughly, and is highly resistant to attempts to circumvent it. It integrates easily with Cursor or via API »

« A model that is 30% faster and up to 30% less expensive than the Sonnet 5 version, designed for debugging, drafting polished documents, and creating presentations. It also allows you to work on texts, interfaces, and images »

« An upgraded version that comes close to the intelligence of the GPT-6 Astra model at about one-fifth the cost. Enjoy better results in programming, writing, and agent-based tasks. You’ll also benefit from a context size exceeding one million tokens »