Anthropic (Claude AI) Statistics (2025)

As of May 2024, Anthropic was valued at approximately $18.4 billion.

 In 2024, Anthropic reported $850 million in annualized revenue and projected it would reach $2.2 billion in 2025.

Anthropic’s estimated revenue grew from $3.4 million in 2021 to $7.9 million in 2022, reaching $45 million in 2023.

Anthropic.com recorded over 7.8 million visits in September 2024.

As of December 2024, 26.26% of Claude AI's desktop traffic originated from the U.S., followed by 8.3% from India and 4.74% from the UK.

As of December 2024, 62.32% of Claude AI's users were male and 37.68% were female.

Claude AI's website attracted approximately 87.6 million monthly visits as of December 2024.

Users viewed an average of 2.72 pages per visit and spent 2 minutes 25 seconds on the site as of September 2024.

In 2024, YouTube accounted for 48.96% of all social media traffic to Claude AI's site.

Claude AI became available in 159 countries globally by the end of 2024.

Anthropic was founded in 2021 by former OpenAI research executives, including Dario Amodei, Ilya Sutskever, and Greg Brockman.

A 2024 research paper estimated Claude AI to have approximately 175 billion parameters.

Claude 3 Haiku was launched on March 13, 2024.

An updated version of Claude 3.5 Haiku was released on October 22, 2024.

Between September and November 2024, Claude AI’s U.S. website ranking improved from 779 to 649.

In the U.S. Programming and Developer Software category, Claude AI’s rank rose from 35 to 27 between September and November 2024.

In 2024, Anthropic was the second-highest funded AI startup globally, having raised $4.2 billion.

The bounce rate for Claude AI's site was 32.25% in November 2024.

In November 2024, users spent an average of 6 minutes on the Claude AI website, viewing 3.73 pages per visit.

As of November 2024, 75.94% of traffic to Claude AI came from direct visits, and 18.57% from organic searches.

In 2024, 49.51% of Claude AI’s social media traffic came from YouTube, followed by WhatsApp Web (12.35%) and Facebook (11.99%).

94.56% of Claude AI’s desktop search traffic came from organic keywords, and 5.44% from paid keywords.

As of December 2024, the keyword “Claude” drove 4.3 million organic visits to the site.

53.68% of Claude AI traffic came from mobile devices, with 46.32% from desktops.

Claude had 18.9 million monthly active users worldwide as of early 2025.

Claude’s mobile app had 2.9 million monthly active users in January 2025.

Claude recorded 16 million unique website visitors in January 2025, down from 18.8 million in November 2024.

Claude’s website traffic grew from 4 million users in December 2023 to a peak of 18.8 million in November 2024.

The United States and India together made up 33.13% of Claude’s total website user base in January 2025.

Claude had 3.2 million website users from the United States in January 2025.

Claude had 2.1 million website users from India in January 2025.

Claude had approximately 692,200 website users from Japan in January 2025.

Claude had around 556,700 website users from South Korea in January 2025.

Claude saw about 523,600 website users from Kenya in January 2025.

Claude reached 516,800 website users in the United Kingdom in January 2025.

Claude recorded 491,100 website users from Germany in January 2025.

Claude had 453,500 website users from Indonesia in January 2025.

Claude attracted 329,200 website users from Israel in January 2025.

Claude had 327,600 website users from Brazil in January 2025.

As of early 2025, 77.1% of Claude’s users were male, while 22.9% were female.

Claude was downloaded 769,600 times in January 2025 across the App Store and Google Play.

Anthropic reported $850 million in annualized revenue and projected $2.2 billion in revenue for 2025.

All Claude 3 models, including Sonnet, Opus, and Haiku, support a 200,000-token context window and can handle inputs over 1 million tokens.

Claude 3 could analyze entire books like The Great Gatsby in one second with perfect recall.

The Claude 3 family includes a Message Batches API that processes up to 10,000 queries per batch at 50% lower cost.

Claude 3 Opus scored 86.8% on the MMLU reasoning test.

Claude 3 Opus scored 161 on the LSAT, 85% on MBE, and 159/166/5.0 on GRE quantitative, verbal, and writing sections.

Claude 3 Opus achieved an average recall of 99.4% across all context lengths.

Claude 3 Sonnet achieved 95.4% accuracy across all context lengths and 91.4% for 200k contexts.

Claude 3 Haiku reduced its refusal rate to below 10%, down from Claude 2.1's 25%.

Claude 3 Haiku achieved an estimated accuracy of 95.9% overall and 91.9% for 200k context lengths.

In January 2024, Anthropic raised over $5 million in Series D funding from investors including MIS and Qualcomm.

The bounce rate for Anthropic.com stood at 53.91% in September 2024.

In September 2024, 67.01% of Anthropic.com visitors were male, while 32.99% were female.

The largest demographic of Anthropic.com visitors in September 2024 was aged 25–34, accounting for 38.56% of users.

As of 2024, Anthropic employs between 101 and 250 people.

The average annual salary for software engineers at Anthropic is around $296,000.

Claude 3.5 Sonnet scored 96.4% from a test which checks a model's ability to solve grade school-level math word problems using logic and multi-step reasoning (GSM8K).

Claude 3.5 Sonnet scored 95.2% from a test that measures how well a model can understand and answer questions based on documents like forms and tables (DocVQA).

Claude 3.5 Sonnet scored 94.7% from a test that evaluates how well a model can interpret and reason about labeled science diagrams (AI2D).

Claude 3.5 Sonnet scored 93.1% from a test designed to assess complex reasoning and problem-solving skills across different tasks (BIG-Bench Hard).

Claude 3.5 Sonnet scored 90.8% from a test that checks how accurately a model can understand and answer questions based on visual charts and data (ChartQA).

Claude 3.5 Sonnet achieved a 92.0% score on HumanEval coding tasks, ahead of Claude 3 Opus and GPT-4o.

Amazon completed a $4 billion investment in Anthropic prior to the release of Claude 3 in March 2024.

Claude 3.5 Sonnet was classified under ASL-2 by Anthropic, indicating early signs of potential risks.

Anthropic launched a new feature called "Artifacts" for Claude, enabling a real-time, editable AI coding workspace.

Anthropic published over 15 research papers on AI safety topics, including interpretability, red teaming, and constitutional AI.

Claude AI was trained using a combination of Constitutional AI and Reinforcement Learning from Human Feedback (RLHF).

Anthropic launched Claude 3 on March 4, 2024, claiming it set a new industry benchmark across multiple cognitive tasks.

Claude AI version 2.1 was made available to users on November 21, 2023.

Anthropic released version 2.0 of Claude AI on July 11, 2023, marking a significant overhaul of the model.

The Claude AI model received its first major update to version 1.3 on April 18, 2023.

Anthropic launched its first Claude AI chatbot (version 1.0) on March 14, 2023.