Anthropic LLMs
Developer of the Claude model family with a focus on constitutional AI and safety research.
Anthropic targets long-context tasks and agentic workflows. Claude has been the default coding pick for many teams since Sonnet 3.5, and the Claude 4.x line extends that with stronger tool-use and a 200K base context window. The API distinguishes itself with prompt caching and a separate Messages endpoint — both noticeably cheap once cache hits dominate.
Founded: 2021 · HQ: San Francisco, USA · Docs: docs.anthropic.com ↗
All Anthropic models
| Model | Family | Context | Input $/M | Output $/M | Released | Status |
|---|---|---|---|---|---|---|
| Claude Opus 4.7 | Claude 4 | 200K | $15.00 | $75.00 | 2026-04-22 | active |
| Claude Sonnet 4.6 | Claude 4 | 200K | $3.00 | $15.00 | 2026-03-12 | active |
| Claude Haiku 4.5 | Claude 4 | 200K | $1.00 | $5.00 | 2025-10-01 | active |
| Claude 3.5 Sonnet | Claude 3 | 200K | $3.00 | $15.00 | 2024-10-22 | active |
Comparisons with other providers
The most-searched comparisons involving Anthropic models:
- Claude Opus 4.7 vs GPT-5 Nano
- Claude Opus 4.7 vs GPT-4o Mini
- Claude Sonnet 4.6 vs GPT-5 Nano
- Claude Sonnet 4.6 vs GPT-4o Mini
- Claude Haiku 4.5 vs GPT-5 Nano
- Claude Haiku 4.5 vs GPT-4o Mini
- Claude 3.5 Sonnet vs GPT-5 Nano
- Claude 3.5 Sonnet vs GPT-4o Mini
Working with the Anthropic API
Documentation lives at https://docs.anthropic.com. Before paying for any call, count input tokens with the appropriate counter:
- OpenAI Token Counter — for GPT-* and o-series models
- Claude Token Counter — for Anthropic Claude models
- Gemini Token Counter — for Google Gemini models