Anthropic Launches Claude Sonnet 5 as a Cheaper Way to Run Agents Anthropic has introduced Claude Sonnet 5, a new midsize model designed to offer agentic capabilities at a lower cost compared to its predecessors. The release marks a shift in the competitive landscape of foundation models, where agentic functionality—allowing systems to plan, use tools, and operate autonomously—is now a baseline expectation. Sonnet 5 is positioned as a more affordable alternative to Anthropic’s larger Opus 4.8 model, with pricing adjustments set to take effect starting June 30, 2026. The model’s agentic features are highlighted as a key differentiator. Anthropic claims Sonnet 5 can perform tasks that previously required more expensive, larger models, such as autonomous planning and tool usage. This aligns with trends observed in other major AI companies, including OpenAI and Google, which have also emphasized agentic capabilities in their recent releases. OpenAI’s GPT-5.6 Sol and Google’s Gemini 3.5 Flash are similarly framed as tools that reduce human oversight in complex workflows. Sonnet 5’s pricing structure is a significant factor in its appeal. Initially priced at $2 per million input tokens and $10 per million output tokens through August 31, 2026, the model becomes more expensive afterward, with input token costs rising to $3 per million. This pricing makes Sonnet 5 cheaper than Opus 4.8, GPT-5.5, and Gemini 3.1 Pro, though it remains more costly than Gemini 3.5 Flash. Anthropic emphasizes that the model’s affordability, combined with its performance, provides developers with a flexible option for balancing cost and efficiency. Performance benchmarks indicate that Sonnet 5 outperforms its predecessor, Sonnet 4.6, in several areas. On agentic coding tasks, Sonnet 5 scores 63.2%, compared to Opus 4.8’s 69.#anthropic #opus_4_8 #claude_sonnet_5 #gpt_5_6_sol #gemini_3_5_flash
Introducing Claude Sonnet 5 Anthropic has launched Claude Sonnet 5, the latest iteration of its agentic AI model series, designed to deliver advanced capabilities in planning, tool integration, and autonomous execution at a lower cost than previous models. This release marks a significant step forward in the agentic AI era, bridging the gap between Sonnet-class and Opus-class models while offering competitive pricing and improved performance. Sonnet 5 is positioned as a substantial upgrade over its predecessor, Sonnet 4.6, with notable improvements in reasoning, tool use, coding, and knowledge work. Performance evaluations show it closely rivals Opus 4.8, a more broadly capable model, but at a significantly lower cost. The model’s agentic capabilities are highlighted through comparisons with earlier versions, demonstrating its ability to handle complex tasks such as multi-step software engineering workflows, cybersecurity assessments, and automated workflows with greater efficiency. Pricing for Sonnet 5 is structured to make it accessible across all user tiers. It is the default model for Free and Pro plans and is available to Max, Team, and Enterprise users. During its introductory launch period through August 31, 2026, Sonnet 5 is priced at $2 per million input tokens and $10 per million output tokens, after which the rates will increase to $3 per million input tokens and $15 per million output tokens. Developers can access the model via the Claude API using the identifier claude-sonnet-5. The model’s performance is benchmarked against Sonnet 4.6 and Opus 4.8 in evaluations such as agentic search (BrowseComp) and computer use (OSWorld-Verified). Sonnet 5 is described as a strict improvement over Sonnet 4.6, offering a balance of cost and performance that makes it a compelling alternative to Opus 4.#anthropic #mythos_preview #claude_sonnet_5 #sonnet_4_6 #opus_4_8
