0

Anthropic

Frontier AI safety lab building the Claude family of models, founded by former OpenAI researchers.

Visit
Type
frontier lab
HQ
San Francisco, CA, USA
Founded
2021
Funding
Private, ~$300B+ valuation 2025; backed by Google, Amazon, Lightspeed, Menlo
Website
anthropic.com
GitHub
github.com/anthropics
Twitter
twitter.com/AnthropicAI

Cite

Notes

Only stored in your browser.

Evals
3
Tools
1
Models
27
Papers
1
Boards
0
People
47

Models

27

Ranked by Artificial Analysis Intelligence Index

#ModelReleasedIntelligence
1Claude Opus 4.828 May 202661.4
2Claude Opus 4.716 Apr 202651.8
3Claude Opus 4.65 Feb 202646.5
4Claude Opus 4.524 Nov 202543.1
5Claude Sonnet 4.617 Feb 202642.6
6Claude Sonnet 4.529 Sep 202537.1
7Claude 4.1 Opus5 Aug 202536
8Claude 4 Sonnet22 May 202533
9Claude 4 Opus22 May 202533
10Claude 4.5 Haiku15 Oct 202531
11Claude Sonnet 3.724 Feb 202530.8
12Claude 3.5 Haiku22 Oct 202418.7
13Claude Opus 34 Mar 202418
14Claude 3.5 Sonnet (Oct '24)22 Oct 202415.9
15Claude 3.5 Sonnet (June '24)21 Jun 202414.2
16Claude 3 Haiku4 Mar 202412.3
17Claude Sonnet 34 Mar 202410.3
18Claude 2.121 Nov 20239.3
19Claude 2.011 Jul 20239.1
20Claude Instant14 Mar 20237.4
21Claude 3.5 Sonnet--
22Claude Mythos Preview--
23Claude Opus 4.5 Agent--
24Claude Sonnet 3.5--
25Claude 3 5 Sonnet--
26Claude Haiku 4 5--
27Claude Opus 4 1--
27 / 27 models

Evals

3

Sycophancy Eval

Evaluate sycophancy of language models across a variety of free-form text-generation tasks.

ActiveAssistants

Anthropic HH-RLHF

Anthropic's helpful & harmless preference dataset - paired human-rated assistant responses widely used both as a preference-training corpus and as a reward-model benchmark.

ActiveSafetyHarmful ContentInstruction Following

Agentic Misalignment: How LLMs could be insider threats

Eliciting unethical behaviour (most famously blackmail) in response to a fictional company-assistant scenario where the model is faced with replacement.

ActiveScheming

Tools

1

Anthropic HH-RLHF

Anthropic's foundational helpful-and-harmless human preference dataset - the first major public RLHF corpus and a long-time community baseline.

PreferenceSafetyJailbreak ResistanceMulti Turn Dialog

Papers

1

Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback

preprint

2022