Anthropic
- Founded: 2021
- Founders: Dario Amodei, Daniela Amodei, Jared Kaplan, Jack Clark, Ben Mann
- Headquarters: San Francisco, California, U.S.
- Website: anthropic.com
Pioneers in Safe and Responsible AI
Anthropic was founded in January 2021 by a group of former OpenAI researchers, led by Dario Amodei and Daniela Amodei. The company was created with the mission of developing safe, beneficial, and understandable artificial intelligence.
Anthropic's founders aim to develop "constitutional AI" systems that are safe, honest, and beneficial. Their intention is to:
- Advance AI safety research
- Create systems that align with fundamental human values
- Develop AI that benefits humanity while minimizing potential risks
- Promote a responsible approach to advanced AI development
Anthropic has secured significant investments from companies like Google and Amazon to continue their research in safe and responsible AI, positioning itself as one of the leading companies in the ethical development of advanced AI systems.
Claude 3.7 Sonnet with Advanced Reasoning
Anthropic, a leading AI safety and research company, has recently unveiled Claude 3.7 Sonnet, the latest iteration in its Claude series of language models. Released on February 24, 2025, Claude 3.7 Sonnet introduces a pioneering hybrid reasoning system that allows users to select between rapid responses and more comprehensive, step-by-step reasoning. This flexibility enhances its performance across a variety of tasks, including programming, complex problem-solving, and creative endeavors.
One of the standout features of Claude 3.7 Sonnet is its "extended thinking mode," which offers deeper analytical capabilities for intricate tasks. Developers also have the ability to adjust the model's "thinking budget," balancing speed and accuracy based on specific requirements.
In addition to Claude 3.7 Sonnet, Anthropic has introduced Claude Code, an agentic command-line tool that enables developers to delegate coding tasks directly from their terminal. This tool aims to streamline coding workflows by allowing Claude to autonomously execute complex, multi-step tasks across various applications.
Claude 3.7 Sonnet is accessible via multiple platforms, including the Claude app, Anthropic's API, Amazon Bedrock, and Google Cloud's Vertex AI.
This release underscores Anthropic's commitment to integrating advanced reasoning capabilities within standard AI models, aiming to enhance user experience and application versatility.
Claude Sonnet 4 launched May, 2025
April, 28 Anthropic, Claude: Web search is now available.
Claude Launch Timeline
- 2023 April: Claude 1.0, el primer asistente de IA disponible públicamente.
- 2023 July: Claude 2.0, con mejoras significativas en razonamiento y capacidades.
- 2023 November: Claude 2.1, con ventana de contexto mejorada y reducción de alucinaciones.
- 2024 March: Familia Claude 3 (Opus, Sonnet y Haiku) enfocada en tareas complejas y velocidad.
- 2024 June - October: Generación Claude 3.5 (Sonnet y Haiku), con salto drástico en programación y visión.
- 2025 February: Claude 3.7 Sonnet, introduciendo capacidades nativas de "Computer Use" e integración híbrida.
- 2025 May: Lanzamiento oficial de Claude 4 (Opus 4 y Sonnet 4), diseñados específicamente para la ejecución de agentes autónomos.
- 2025 Late (Aug - Nov): Lanzamientos de la subfamilia de optimización incremental: Claude Opus 4.1, Sonnet 4.5 y Haiku 4.5.
- 2026 February: Claude Sonnet 4.6 e Opus 4.6, con ventana de contexto extendida a 1 millón de tokens (beta), razonamiento adaptativo y la suite de herramientas colaborativas "Cowork".
- 2026 April: Claude Opus 4.7 (actual modelo de referencia en el mercado) junto con la plataforma multimedia Claude Design.
- 2026 May (Novedad Reciente): Despliegue de Claude Mythos Preview como parte del macroproyecto de seguridad defensiva Project Glasswing.
