Claude 3
Anthropic's third-generation model family released in March 2024, featuring three tiers: Haiku (fast/cheap), Sonnet (balanced), and Opus (most capable). Claude 3 introduced multimodal vision capabilities, significantly improved performance, and larger context windows up to 200K tokens.
Implements
Concepts this tool claims to implement:
- Large Language Model primary
State-of-the-art language models competitive with GPT-4 across reasoning, coding, and knowledge tasks.
- Multimodal Model primary
All Claude 3 models support image understanding and analysis.
- Chat Completions API primary
Messages API following chat completions pattern with system/user/assistant roles.
- Function Calling primary
Tool use support for structured function calling and agent applications.
Integration Surfaces
Details
- Vendor
- Anthropic
- License
- Proprietary
- Runs On
- cloud
- Used By
- human, agent, system
Links
Notes
Claude 3 established Anthropic as a serious competitor in the frontier model space. The three-tier structure (Haiku/Sonnet/Opus) became a common pattern for offering models at different capability/cost tradeoffs.