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
24
Papers
1
Boards
0
People
47

Models

24

Ranked by Artificial Analysis Intelligence Index

#ModelReleasedIntelligence
1Claude Fable 59 Jun 202659.9
2Claude Opus 4.828 May 202655.7
3Claude Sonnet 530 Jun 202653.4
4Claude Opus 4.716 Apr 202642.7
5Claude Sonnet 5 (Non-reasoning, High Effort)30 Jun 202641.7
6Claude Opus 4.65 Feb 202637.8
7Claude Opus 4.524 Nov 202534.7
8Claude Sonnet 4.617 Feb 202634.3
9Claude Sonnet 4.529 Sep 202529.3
10Claude 4.1 Opus5 Aug 202528.2
11Claude 4 Sonnet22 May 202525.5
12Claude 4 Opus22 May 202525.5
13Claude 4.5 Haiku15 Oct 202523.7
14Claude Sonnet 3.724 Feb 202523.5
15Claude 3.5 Haiku22 Oct 202412.3
16Claude Opus 34 Mar 202411.8
17Claude 3.5 Sonnet22 Oct 20249.9
18Claude 3.5 Sonnet (June '24)21 Jun 20248.3
19Claude Sonnet 34 Mar 20244.7
20Claude 3 Haiku4 Mar 20243.9
21Claude 2.121 Nov 20233.9
22Claude 2.011 Jul 20233.6
23Claude Instant14 Mar 20232.1
24Claude Mythos Preview--
24 / 24 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