Anthropic has released a new version of its AI model, Claude 3.5 Haiku, for users of the Claude chatbot platform. The model is now available on the website and in mobile apps. Announcements about the launch of 3.5 Haiku began appearing on social media on Thursday morning.
Claude 3.5 Haiku, introduced in November, matches or surpasses the performance of Anthropic’s previous flagship model, 3 Opus, in certain benchmarks. Anthropic notes that 3.5 Haiku is particularly effective for coding recommendations, data extraction and labeling, as well as content moderation. The model can generate longer text fragments than its predecessor, 3 Haiku, and features an updated knowledge base, allowing it to reference more current events.
However, 3.5 Haiku does not support image analysis, making it less powerful in this aspect compared to other Anthropic models such as 3 Haiku and 3.5 Sonnet. The model became the subject of a minor controversy when it appeared in the Anthropic API at the beginning of last month. Initially, the company stated that 3.5 Haiku would cost the same as 3 Haiku, but later changed its position, arguing that the model’s increased “intelligence” justified a higher API price.
