Claude Blog 采集 (2026-06-02)¶
共采集 5 篇文章
📋 文章索引¶
- The advisor strategy: Give agents an intelligence boost - Apr 09, 2026 (评分: 9.0)
- The founder's playbook: Building an AI-native startup - May 14, 2026 (评分: 9.0)
- Token-saving updates on the Anthropic API - Mar 13, 2025 (评分: 9.0)
- Claude can now use tools - May 30, 2024 (评分: 9.0)
- Claude 3.5 Haiku on AWS Trainium2 and model distillation in Amazon Bedrock - Dec 03, 2024 (评分: 9.0)
The advisor strategy: Give agents an intelligence boost¶
来源: Claude Blog 发布日期: Apr 09, 2026 采集时间: 2026-06-02 价值评分: 9.0/10 正文字数: ~2493 字符
摘要¶
Pair Opus as an advisor with Sonnet or Haiku as an executor, and get Opus-level intelligence in your agents at a fraction of the cost.
正文内容¶
Pair Opus as an advisor with Sonnet or Haiku as an executor, and get near Opus-level intelligence in your agents at a fraction of the cost.
Developers who want to better balance intelligence and cost have converged on what we call the advisor strategy: pair Opus as an advisor with Sonnet or Haiku as an executor. This brings near Opus-level intelligence to your agents while keeping costs near Sonnet levels. Today we're introducing the advisor tool on the Claude Platform to make the advisor strategy a one-line change in your API call. Build cost-effective agents with the advisor strategy
Get started The advisor tool is available now in beta natively on the Claude Platform. To get started: Add the beta feature header: anthropic-beta: advisor-tool-2026-03-01 Add the advisor_20260301 to your Messages API request Modify your system prompt based on your use case We recommend running your existing eval suite against Sonnet solo, Sonnet executor with Opus advisor, and Opus solo. Explore the docs to learn more. Footnotes SWE-bench Multilingual: Sonnet 4.6 solo used adaptive thinking. Sonnet 4.6 + Advisor used our suggested system prompt for coding with thinking turned off. Both runs used high effort with bash and file editing tools. Scores are averaged over five trials of 300 problems across nine languages. Opus 4.6 was used as the advisor model in all runs. BrowseComp: All runs used thinking turned off with web search and web fetch tools. Sonnet 4.6 runs used medium effort. Sonnet 4.6 + Advisor used our suggested system prompt for coding; Haiku 4.5 + Advisor did not. No programmatic tool calling or context compaction. Scores are based on 1,266 problems with one attempt per problem. Opus 4.6 was used as the advisor model in all runs. Terminal-Bench 2.0: All runs used thinking turned off with bash and file editing tools. Sonnet 4.6 runs used medium effort. Neither advisor run used our suggested system prompt for coding. Each task ran in an isolated pod with 3x resource allocation and a 1x timeout. Scores are averaged over five attempts per task across 89 tasks. Opus 4.6 was used as the advisor model in all runs.
Explore more product news and best practices for teams building with Claude.
Transform how your organization operates with Claude
Product updates, how-tos, community spotlights, and more. Delivered monthly to your inbox.
Please provide your email address if you'd like to receive our monthly developer newsletter. You can unsubscribe at any time.
采集自 Claude Blog，由 collect_claude_blog.py 自动采集
The founder's playbook: Building an AI-native startup¶
来源: Claude Blog 发布日期: May 14, 2026 采集时间: 2026-06-02 价值评分: 9.0/10 正文字数: ~1996 字符
摘要¶
We share how AI-native founders are using Claude at every stage of the startup journey, with practical exercises, frameworks, and prompts.
正文内容¶
We share how founders are using AI at every stage of the startup journey, with practical exercises, frameworks, and prompts for using Claude.
AI is reshaping how startups are being built. Founders who've never written a line of code before are shipping production applications, reaching revenue before scaling headcount, and building tools to automate their most tedious workflows. The founder's role is shifting from individual contributor to orchestrator, allowing them to focus on the work only they can do. We put together a practical playbook for building an AI-native startup. It remaps the four core stages of the startup lifecycle—Idea, MVP, Launch, and Scale—for what's possible in 2026, with the goals, exit criteria, common failure modes, and AI-powered exercises that work at each one. In this playbook, we share: How to validate a problem hypothesis, map a competitive landscape, and run customer discovery with AI Architecture, scope, and security practices that keep AI-generated MVP codebases from accruing technical debt A measurement framework for distinguishing genuine product-market fit from early hype A Launch-stage operating system that replaces founder attention with agentic workflows A product matrix for when and how to use Chat, Claude Cowork, and Claude Code across each stage of the startup journey Founder stories from Ambral, Anything, Carta Healthcare, HumanLayer, Vulcan Technologies, and more These best practices were written for founders deciding how to architect their company around AI from day one and for the early operators helping them get there. Check it out, here . Get started with Claude today.
Explore more product news and best practices for teams building with Claude.
Transform how your organization operates with Claude
Product updates, how-tos, community spotlights, and more. Delivered monthly to your inbox.
Please provide your email address if you'd like to receive our monthly developer newsletter. You can unsubscribe at any time.
采集自 Claude Blog，由 collect_claude_blog.py 自动采集
Token-saving updates on the Anthropic API¶
来源: Claude Blog 发布日期: Mar 13, 2025 采集时间: 2026-06-02 价值评分: 9.0/10 正文字数: ~1721 字符
摘要¶
Claude now offers cache-aware rate limits, simplified prompt caching, and token-efficient tool use to help developers increase throughput and cut costs. We've made several updates to the Anthropic API that let developers significantly increase throughput and reduce token usage with Claude 3.
正文内容¶
Claude now offers cache-aware rate limits, simplified prompt caching, and token-efficient tool use to help developers increase throughput and cut costs.
We've made several updates to the Anthropic API that let developers significantly increase throughput and reduce token usage with Claude 3.7 Sonnet. These include: cache-aware rate limits, simpler prompt caching, and token-efficient tool use. Together, these updates will help you process more requests within your existing rate limits and reduce costs with minimal code changes. Increase your throughput with prompt caching Prompt caching allows developers to store and reuse frequently accessed context between API calls. This lets Claude maintain knowledge of large documents, instructions, or examples without sending the same information with each request—reducing costs by up to 90% and latency by up to 85% for long prompts. We’ve released two improvements to prompt caching for Claude 3.7 Sonnet that work together to help you scale more efficiently. Cache-aware rate limits Prompt cache read tokens no longer count against your Input Tokens Per Minute (ITPM) limit for Claude 3.7 Sonnet on the Anthropic API. This means you can now optimize your prompt caching usage to increase throughput and get more out of your existing ITPM rate limits. Your Output Tokens Per Minute (OTPM) rate limit remains the same.
Explore more product news and best practices for teams building with Claude.
Transform how your organization operates with Claude
Product updates, how-tos, community spotlights, and more. Delivered monthly to your inbox.
Please provide your email address if you'd like to receive our monthly developer newsletter. You can unsubscribe at any time.
采集自 Claude Blog，由 collect_claude_blog.py 自动采集
Claude can now use tools¶
来源: Claude Blog 发布日期: May 30, 2024 采集时间: 2026-06-02 价值评分: 9.0/10 正文字数: ~1669 字符
摘要¶
Claude now connects with external tools and APIs to perform tasks, manipulate data, and deliver more accurate responses. Tool use, which enables Claude to interact with external tools and APIs, is now generally available across the entire Claude 3 model family on the Anthropic Messages API, Amazon...
正文内容¶
Claude now connects with external tools and APIs to perform tasks, manipulate data, and deliver more accurate responses.
Tool use, which enables Claude to interact with external tools and APIs, is now generally available across the entire Claude 3 model family on the Anthropic Messages API, Amazon Bedrock, and Google Cloud's Vertex AI. With tool use, Claude can perform tasks, manipulate data, and provide more dynamic—and accurate—responses. Tool use Define a toolset for Claude and specify your request in natural language. Claude will then select the appropriate tool to fulfill the task and, when appropriate, execute the corresponding action: Extract structured data from unstructured text : Pull names, dates, and amounts from invoices to reduce manual data entry. Convert natural language requests into structured API calls : Enable teams to self-serve common actions (e.g., "cancel subscription") with simple commands. Answer questions by searching databases or using web APIs : Provide instant, accurate responses to customer inquiries in support chatbots. Automate simple tasks through software APIs : Save time and minimize errors in data entry or file management. Orchestrate multiple fast Claude subagents for granular tasks : Automatically find the optimal meeting time based on attendee availability.
Explore more product news and best practices for teams building with Claude.
Transform how your organization operates with Claude
Product updates, how-tos, community spotlights, and more. Delivered monthly to your inbox.
Please provide your email address if you'd like to receive our monthly developer newsletter. You can unsubscribe at any time.
采集自 Claude Blog，由 collect_claude_blog.py 自动采集
Claude 3.5 Haiku on AWS Trainium2 and model distillation in Amazon Bedrock¶
来源: Claude Blog 发布日期: Dec 03, 2024 采集时间: 2026-06-02 价值评分: 9.0/10 正文字数: ~4173 字符
摘要¶
We're bringing faster inference to Claude 3.5 Haiku through AWS Trainium2 optimization and enabling model distillation in Amazon Bedrock to help you achieve frontier-level performance at lower costs.
正文内容¶
We're bringing faster inference to Claude 3.5 Haiku through AWS Trainium2 optimization and enabling model distillation in Amazon Bedrock to help you achieve frontier-level performance at lower costs.
As part of our expanded collaboration with AWS , we’ve begun optimizing Claude models to run on AWS Trainium2 , their most advanced AI chip. To preview what’s possible with Trainium2, Claude 3.5 Haiku now supports latency-optimized inference in Amazon Bedrock , making the model significantly faster without compromising accuracy. We’re also adding support for model distillation in Amazon Bedrock, bringing the intelligence of larger Claude models to our faster and more cost-effective models. Next-gen models on Trainium2 We are collaborating with AWS to build Project Rainier—an EC2 UltraCluster of Trn2 UltraServers containing hundreds of thousands of Trainium2 chips. This cluster will deliver more than five times the computing power (in exaflops) used to train our current generation of leading AI models. Trainium2 enables us to offer faster models in Amazon Bedrock, starting with Claude 3.5 Haiku which now supports latency-optimized inference in public preview. By enabling latency optimization, Claude 3.5 Haiku can deliver up to 60% faster inference speed—making it the ideal choice for use cases ranging from code completions to real-time content moderation and chatbots. This faster version of Claude 3.5 Haiku, powered by Trainium2, is available in the US East (Ohio) Region via cross-region inference and is offered at $1 per million input tokens and $5 per million output tokens. Amazon Bedrock Model Distillation We’re also enabling customers to get frontier performance from Claude 3 Haiku—our most cost-effective model from the last generation. With distillation, Claude 3 Haiku can now achieve significant performance gains, reaching Claude 3.5 Sonnet-like accuracy for specific tasks—at the same price and speed of our most cost-effective model. This technique transfers knowledge from the "teacher" (Claude 3.5 Sonnet) to the "student" (Claude 3 Haiku), enabling customers to run sophisticated tasks like retrieval augmented generation (RAG) and data analysis at a fraction of the cost. Unlike traditional fine-tuning, which requires developers to manually craft training examples and continuously adjust parameters, Amazon Bedrock Model Distillation automates the entire process by: Generating synthetic training data from Claude 3.5 Sonnet Training and evaluating Claude 3 Haiku Hosting the final distilled model for inference Amazon Bedrock Model Distillation automatically applies different data synthesis methods—from generating similar prompts to creating new high-quality responses based on your example prompt-response pairs. Distillation for Claude 3 Haiku in Amazon Bedrock is now available in preview. Learn more in the AWS launch blog and documentation . Lower prices for Claude 3.5 Haiku In addition to offering a faster version on Trainium2, customers can continue to access Claude 3.5 Haiku on the Anthropic API , Amazon Bedrock , and Google Cloud’s Vertex AI . To make this model even more accessible for a wide range of use cases, we’re lowering the price of Claude 3.5 Haiku to $0.80 per million input tokens and $4 per million output tokens across all platforms. Get started Starting today, model distillation and the faster Claude 3.5 Haiku are available in preview in Amazon Bedrock. For developers seeking the optimal balance of price, performance, and speed, you now have expanded model options with Claude: Claude 3.5 Haiku with latency optimization, powered by Trainium2, for general use cases Claude 3 Haiku, distilled with frontier performance, for high-volume, repetitive use cases To get started, visit the Amazon Bedrock console . We can’t wait to see what you build.
Explore more product news and best practices for teams building with Claude.
Transform how your organization operates with Claude
Product updates, how-tos, community spotlights, and more. Delivered monthly to your inbox.
Please provide your email address if you'd like to receive our monthly developer newsletter. You can unsubscribe at any time.
采集自 Claude Blog，由 collect_claude_blog.py 自动采集