Artificial intelligence (AI) company Anthropic released its latest chatbot, Claude 3.5 Sonnet, on Thursday (June 20). The bot supposedly outperforms its previous top-tier offering, Claude 3 Opus. Claude 3.5 Sonnet is now available for free on Claude.ai and the Claude iOS app, with higher usage limits for subscribers to Claude Pro and Team plans. The […]
The post Anthropic Debuts New AI Model, Claude 3.5 Sonnet, That Understands Humor appeared first on PYMNTS.com.
Artificial intelligence (AI) company Anthropic released its latest chatbot, Claude 3.5 Sonnet, on Thursday (June 20). The bot supposedly outperforms its previous top-tier offering, Claude 3 Opus.
Claude 3.5 Sonnet is now available for free on Claude.ai and the Claude iOS app, with higher usage limits for subscribers to Claude Pro and Team plans. The model can also be accessed through Anthropic’s API, Amazon Bedrock and Google Cloud’s Vertex AI.
The company said in a blog post that Claude 3.5 Sonnet “sets new industry benchmarks for graduate-level reasoning (GPQA), undergraduate-level knowledge (MMLU), and coding proficiency (HumanEval).” Anthropic notes improvements in the model’s ability to grasp “nuance, humor, and complex instructions,” and describes it as “exceptional at writing high-quality content with a natural, relatable tone.”
A key feature of Claude 3.5 Sonnet is its processing speed, which Anthropic claims is twice that of Claude 3 Opus. This speed boost, combined with what the company calls “cost-effective pricing,” positions the model for complex tasks such as “context-sensitive customer support and orchestrating multi-step workflows.”
In an internal evaluation of agentic coding, the company said, Claude 3.5 Sonnet solved 64% of problems, compared to 38% for Claude 3 Opus. The test assessed the model’s ability to “fix a bug or add functionality to an open-source codebase, given a natural language description of the desired improvement.”
Anthropic also touts Claude 3.5 Sonnet as its “strongest vision model yet,” surpassing Claude 3 Opus on standard vision benchmarks. The company highlights improvements in visual reasoning tasks, such as interpreting charts and graphs, and the ability to “accurately transcribe text from imperfect images.”
Alongside the model release, Anthropic introduced a new feature called Artifacts on Claude.ai. This addition creates a dedicated window for AI-generated content like code snippets and text documents. The company describes this as “a dynamic workspace where [users] can see, edit, and build upon Claude’s creations in real-time.”
Anthropic has been working overtime to add new AI features to the chatbot race. As previously reported by PYMNTS, Claude AI now boasts a “tool use” feature, enabling businesses to create custom AI helpers. This upgrade enhances customer support and optimizes operations.
Anthropic emphasized its commitment to safety and privacy, stating that Claude 3.5 Sonnet has undergone “rigorous testing” and has been “trained to reduce misuse.” The company engaged external experts, including the U.K.’s Artificial Intelligence Safety Institute (UK AISI), for pre-deployment safety evaluation.
The release notes that Anthropic incorporated “policy feedback from outside subject matter experts to ensure that our evaluations are robust and take into account new trends in abuse.” This included input from child safety experts at Thorn to update classifiers and fine-tune the models.
Anthropic reaffirmed its stance on user privacy, stating, “We do not train our generative models on user-submitted data unless a user gives us explicit permission to do so.”
Looking ahead, Anthropic plans to release Claude 3.5 Haiku and Claude 3.5 Opus later this year to complete the Claude 3.5 model family. The company is also developing new features and integrations, including a Memory feature that will enable Claude to “remember a user’s preferences and interaction history as specified.”
For all PYMNTS AI coverage, subscribe to the daily AI Newsletter.
The post Anthropic Debuts New AI Model, Claude 3.5 Sonnet, That Understands Humor appeared first on PYMNTS.com.
