Anthropic announced on April 9 that it would正式地正式引入 the “Advisor Strategy” into the Claude Platform. Developers can now have Opus play an advisory role through the Messages API, pairing Sonnet or Haiku as the executor, achieving near-Opus-level intelligence in a single API request while significantly lowering costs. This capability is currently offered in the form of a public beta.
How it works: executors automatically consult Opus when they encounter difficult problems
The core idea behind the Advisor Strategy is to separate reasoning from execution. After developers add an advisor tool in their Messages API call, when a Sonnet or Haiku agent faces a difficult decision during task execution, it will automatically consult Opus, obtain planning guidance, and then continue executing. The entire process is completed within a single API request, with no extra orchestration required from the developer.
This architecture is similar to the division of labor between senior consultants and operators in a team—Opus handles high-level strategic judgment, while Sonnet or Haiku handles actual operations. The two collaborate within the same workflow.
Hands-on results: better performance at lower cost
According to the evaluation data Anthropic published, on the SWE-bench Multilingual benchmark, Sonnet combined with the Opus advisor performed 2.7 percentage points better than using Sonnet alone, while the cost per task actually dropped by 11.9%.
The reason costs decreased is that most tasks are still handled by the cheaper Sonnet or Haiku. Only at critical decision points is Opus called, avoiding the high expense of using Opus throughout the entire process.
What it means for developers
The Advisor Strategy provides developers building AI agents with a new cost-efficiency trade-off option. In the past, developers had to choose between “using Opus end-to-end but at high cost” or “using Sonnet end-to-end, but risking mistakes when encountering complex problems.” Now, it’s possible to get the advantages of both through a hybrid architecture.
The community’s reaction has been enthusiastic. Some developers likened it to the “call for help” lifeline from the “Millionaire” TV show—executors work independently most of the time, and only turn to the experts when they encounter truly difficult problems.
This article, Anthropic launches the Advisor Strategy: Opus as an advisor paired with Sonnet for execution, improving performance and reducing costs by 12%, first appeared on 鏈新聞 ABMedia.
Disclaimer: The information on this page may come from third parties and does not represent the views or opinions of Gate. The content displayed on this page is for reference only and does not constitute any financial, investment, or legal advice. Gate does not guarantee the accuracy or completeness of the information and shall not be liable for any losses arising from the use of this information. Virtual asset investments carry high risks and are subject to significant price volatility. You may lose all of your invested principal. Please fully understand the relevant risks and make prudent decisions based on your own financial situation and risk tolerance. For details, please refer to
Disclaimer.
Related Articles
Guo Ming-chi: OpenAI wants to build an AI Agent phone; MediaTek, Qualcomm, and Luxshare Precision are key in the supply chain
Guo Ming-chi claims that OpenAI is working with MediaTek, Qualcomm, and Luxshare Precision to develop an AI Agent phone, with mass production expected in 2028. The new phone will be centered on task completion: an AI agent will understand and execute requests, combining cloud and on-device computing, with a focus on sensing and contextual understanding. The specifications and supply chain list are expected to be finalized in 2026–2027; if it takes shape, it could bring a new upgrade cycle to the high-end market, and Luxshare may become a major beneficiary.
ChainNewsAbmedia4m ago
IEA: AI infrastructure spending has already surpassed investment in oil and gas production, and is expected to increase another 75% in 2026
According to analysis and market data published by the International Energy Agency (IEA) on April 26, the combined capital expenditures of the world’s top five technology companies in 2025 exceed $400 billion, with most of the spending going toward building AI infrastructure. The scale has already surpassed the annual investment level of global oil and natural gas production. The IEA estimates that the related capital expenditures may further increase by 75% in 2026.
MarketWhisper36m ago
Senator Bernie Sanders Issues Warning on AI's Existential Threat
Sanders stressed that even as most AI scientists acknowledge the possibility of AI escaping control and becoming a danger to our existence, no major measures have been taken to avoid it. “We must make certain that Al benefits humanity, not hurts us,” he stated.
Key Takeaways:
Bernie Sanders
Coinpedia45m ago
Xiaomi’s AI model lead: As AI competition shifts to the Agent era, self-evolution is a key event on the path to AGI
Xiaomi’s large-model team head, Luo Fuli, accepted an in-depth interview on the Bilibili platform on April 24 (video ID: BV1iVoVBgERD). The interview lasted 3.5 hours, and it was her first time, as the technical head, to publicly and systematically explain her technical viewpoints. Luo Fuli said that the large-model competition track has shifted from the Chat era to the Agent era, and she pointed out that “self-evolution” will be a key event for AGI in the coming year.
MarketWhisper46m ago
xAI Grok Voice takes over Starlink customer service hotline, with 70% of calls automatically closed
According to an official announcement from xAI released on April 23, xAI has launched the Grok Voice Think Fast 1.0 voice AI agent, and it has already been deployed to the Starlink customer service hotline +1 (888) GO STARLINK. According to the test data disclosed in the announcement, 70% of calls are automatically closed by AI, with no human intervention required.
MarketWhisper57m ago
GPT-5.5 Returns to Cutting Edge in Coding, But OpenAI Switches Benchmarks After Losing to Opus 4.7
Gate News message, April 27 — SemiAnalysis, a semiconductor and AI analysis firm, released a comparative benchmark of coding assistants including GPT-5.5, Claude Opus 4.7, and DeepSeek V4. The key finding: GPT-5.5 marks OpenAI's first return to the cutting edge in coding models in six months, with S
GateNews1h ago
