正在加载视频...

视频加载失败

加载此视频时出现问题。这可能是由于临时网络问题，或视频可能不可用。

Bolt's agent now runs Claude 3.7 Sonnet—a huge upgrade for coding, design & debugging! But that’s not all… We’re introducing Dynamic Reasoning, a first-of-its-kind feature that lets the AI choose how hard to “think” when tackling complex problems.

bolt.new

118,557 subscribers

70,684 次观看 • 1 年前 •via X (Twitter)

Anya Rossi• Live Now

Private livecam show

11 条评论

bolt.new 的头像

bolt.new1 年前

To get started: ✨ Claude 3.7 Sonnet is live on all projects 🧠 Dynamic Reasoning is in preview—enable it in Settings > Feature Previews. Note: While in preview, DR currently only supports smaller projects. Let us know what you think!

FRANK E ELKINS 的头像

FRANK E ELKINS1 年前

Confused about AI? Get clarity today with Book VI - "The Rational Being!" Understand the benefits & risks, empower your future now by learning what AI really is and how it really works! Also check out the Free Weekly Newsletter "How Things Work: A Brief History of Reality"

skippy 的头像

skippy1 年前

here we gooooo this changes the game… again!

Hermes ᯅ 的头像

Hermes ᯅ1 年前

Lets go!! Can't wait to try this out!

Mitchell Wright 🇺🇦 的头像

Mitchell Wright 🇺🇦1 年前

Huge

Sentedz 的头像

Sentedz1 年前

Would like to test but I'm running low on creds 💔

TheHedgehog.eth | NamesPortfolio.com 的头像

TheHedgehog.eth | NamesPortfolio.com1 年前

👏🏼😎🫡

Vedant 的头像

Vedant1 年前

Sonnet is beast for coding Here's how the started

Eddymarc 的头像

Eddymarc1 年前

Is this Claude 3.7 standard or extended for coding???? @ericsimons40 @sulco

Tobasco 的头像

Tobasco1 年前

YES! LFG should help me complete my app been stuck the past couple days. 🫡

Shane Carman 的头像

Shane Carman1 年前

Any plans for use on a mobile phone without crashing?

相关视频

Here’s how to build a Slack clone with Claude 3.7 Sonnet and Cursor Agent in v0.46. Perfect for *anyone* wanting to pick up AI coding tools. Learn how to build an app 100% from scratch with the newest AI coding workflows. Watch for 1hr+ of agentic coding.

Here’s how to build a Slack clone with Claude 3.7 Sonnet and Cursor Agent in v0.46. Perfect for anyone wanting to pick up AI coding tools. Learn how to build an app 100% from scratch with the newest AI coding workflows. Watch for 1hr+ of agentic coding.

Mckay Wrigley

212,484 次观看 • 1 年前

Claude 3.7 reasoning and coding capabilities are no joke! Watch how I use it to one-shot a quick simulator of how attention mechanisms work. I think it could be awesome for each of us to have access to a personal Andrej Karpathy-like tutor to explain complex stuff to us. 😀

Claude 3.7 reasoning and coding capabilities are no joke! Watch how I use it to one-shot a quick simulator of how attention mechanisms work. I think it could be awesome for each of us to have access to a personal Andrej Karpathy-like tutor to explain complex stuff to us. 😀

elvis

31,669 次观看 • 1 年前

AI coding tools are everywhere. Yet they all fall short when scaling to 100s of developers and complex codebases. We’re solving this problem Augment Code. Introducing Augment Code, the first developer AI purpose-built for teams.🧵

AI coding tools are everywhere. Yet they all fall short when scaling to 100s of developers and complex codebases. We’re solving this problem Augment Code. Introducing Augment Code, the first developer AI purpose-built for teams.🧵

Scott Dietzen

91,368 次观看 • 1 年前

The lack of "feature design" is why so many products over time feel hollow or messy. This isn't visual design. This isn't architectural design. I thought that a short video lecture of what feature design is and a real case study of applying it in Ghostty would be helpful. Feature design is the planning step behind how you're going to solve one or more user problems with a product feature: what that feature looks like, how it feels, and not just how its going to tactically solve these specific problems, but how that solution is going to interface with the edges of other features that currently exist or are planned to exist in the future.

The lack of "feature design" is why so many products over time feel hollow or messy. This isn't visual design. This isn't architectural design. I thought that a short video lecture of what feature design is and a real case study of applying it in Ghostty would be helpful. Feature design is the planning step behind how you're going to solve one or more user problems with a product feature: what that feature looks like, how it feels, and not just how its going to tactically solve these specific problems, but how that solution is going to interface with the edges of other features that currently exist or are planned to exist in the future.

Mitchell Hashimoto

156,227 次观看 • 7 个月前

We’re introducing Workspace Intelligence. With the advanced reasoning of Gemini, we're eliminating context silos across your Workspace apps. Think of it as a unified intelligence layer that works across Workspace. See how Workspace Intelligence connects the dots and lets AI do the heavy lifting for you in this demo. #GoogleCloudNext

We’re introducing Workspace Intelligence. With the advanced reasoning of Gemini, we're eliminating context silos across your Workspace apps. Think of it as a unified intelligence layer that works across Workspace. See how Workspace Intelligence connects the dots and lets AI do the heavy lifting for you in this demo. #GoogleCloudNext

Google Workspace

123,846 次观看 • 3 个月前

RouteLLM: Choose the Best LLM for Your Query! Use a single interface to access all state-of-the-art LLMs: - o1 for hard problems - Sonnet 3.5 for standard coding - GPT-4o for essays - Gemini Flash 2.0 for longer contexts Mix and match to use the best LLM for your query on ChatLLM.

RouteLLM: Choose the Best LLM for Your Query! Use a single interface to access all state-of-the-art LLMs: - o1 for hard problems - Sonnet 3.5 for standard coding - GPT-4o for essays - Gemini Flash 2.0 for longer contexts Mix and match to use the best LLM for your query on ChatLLM.

Akshay 🚀

56,316 次观看 • 1 年前

🚨 this chinese guy makes over $1,000,000 a year… by building AI agents. no employees. no massive startup. he just keeps building. while most people are still asking ChatGPT random questions, he’s using Claude to build software that solves real problems. this is what people call vibe coding. he opens Claude and says: “build me an AI agent for real estate businesses that creates property videos.” Claude writes the code. builds the interface. adds subscriptions. helps deploy the app. within a day, he has a working product. then he starts building the next one. that’s the part most people don’t understand. he isn’t trying to build one billion-dollar company. he’s building dozens of AI agents, each solving one problem for one industry. → an AI agent for dentists → an AI agent for ecommerce brands → an AI agent for podcasters → an AI agent for real estate businesses each one automates work that people normally do by hand. each one is built with simple prompts. each one can become a real business. the crazy part? you don’t need to be a software engineer anymore. you need to know how to think like a builder. how to spot problems. how to explain solutions to AI. and how to ship. that’s exactly why i’m reading this article: “How to Actually Build Your First AI Agent.” because this is the skill that’s creating the next generation of builders. the people who learn to build AI agents today won’t just use AI. they’ll own the tools everyone else ends up paying for.

🚨 this chinese guy makes over $1,000,000 a year… by building AI agents. no employees. no massive startup. he just keeps building. while most people are still asking ChatGPT random questions, he’s using Claude to build software that solves real problems. this is what people call vibe coding. he opens Claude and says: “build me an AI agent for real estate businesses that creates property videos.” Claude writes the code. builds the interface. adds subscriptions. helps deploy the app. within a day, he has a working product. then he starts building the next one. that’s the part most people don’t understand. he isn’t trying to build one billion-dollar company. he’s building dozens of AI agents, each solving one problem for one industry. → an AI agent for dentists → an AI agent for ecommerce brands → an AI agent for podcasters → an AI agent for real estate businesses each one automates work that people normally do by hand. each one is built with simple prompts. each one can become a real business. the crazy part? you don’t need to be a software engineer anymore. you need to know how to think like a builder. how to spot problems. how to explain solutions to AI. and how to ship. that’s exactly why i’m reading this article: “How to Actually Build Your First AI Agent.” because this is the skill that’s creating the next generation of builders. the people who learn to build AI agents today won’t just use AI. they’ll own the tools everyone else ends up paying for.

MIKE

38,108 次观看 • 1 个月前

The agents are coming. The real question is: how will we work with them? Common Ground is our answer. Introducing a new way to build, direct, and inspect AI systems. • Full visualization of agent team dynamics • Shared memory across tasks • Transparent reasoning • Runs locally with Gemini-CLI (free) Start collaborating with systems that think with you, not for you.

The agents are coming. The real question is: how will we work with them? Common Ground is our answer. Introducing a new way to build, direct, and inspect AI systems. • Full visualization of agent team dynamics • Shared memory across tasks • Transparent reasoning • Runs locally with Gemini-CLI (free) Start collaborating with systems that think with you, not for you.

Intelligent Internet

325,922 次观看 • 1 年前

$ANTHROPIC JUST QUIETLY SHIPPED A FEATURE THAT LETS CLAUDE SPAWN A WHOLE TEAM OF AGENTS THAT MESSAGE EACH OTHER AND REVIEW EACH OTHER'S WORK. It's a Claude Code feature called agent teams. The team lead spawns multiple agents that share a task list and message each other directly, not subagents reporting back, actual peers. In the demo a QA agent caught three bugs, sent the work back to the front-end and back-end devs, they fixed it, app shipped in one pass. How to run it: 1. Enable it. Needs Claude Code v2.1.32+. Add to settings.json: "env": { "CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS": "1" }. Or paste that to Claude and say "add this to my settings." Restart. 2. Prompt in plain English. Start with a goal (agents wake with zero context), then "create a team of 3 using Sonnet," describe each role, its deliverable, and who it messages when done. 3. The rules: each agent owns its own files, define exact outputs, name who talks to who, keep it to 3-5 agents. Use it for complex work with separate parts running in parallel. Skip it for simple or sequential tasks, teams cost 3-4x the tokens. Bookmark this.$

ANTHROPIC JUST QUIETLY SHIPPED A FEATURE THAT LETS CLAUDE SPAWN A WHOLE TEAM OF AGENTS THAT MESSAGE EACH OTHER AND REVIEW EACH OTHER'S WORK. It's a Claude Code feature called agent teams. The team lead spawns multiple agents that share a task list and message each other directly, not subagents reporting back, actual peers. In the demo a QA agent caught three bugs, sent the work back to the front-end and back-end devs, they fixed it, app shipped in one pass. How to run it: 1. Enable it. Needs Claude Code v2.1.32+. Add to settings.json: "env": { "CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS": "1" }. Or paste that to Claude and say "add this to my settings." Restart. 2. Prompt in plain English. Start with a goal (agents wake with zero context), then "create a team of 3 using Sonnet," describe each role, its deliverable, and who it messages when done. 3. The rules: each agent owns its own files, define exact outputs, name who talks to who, keep it to 3-5 agents. Use it for complex work with separate parts running in parallel. Skip it for simple or sequential tasks, teams cost 3-4x the tokens. Bookmark this.

Yarchi

460,553 次观看 • 1 个月前

Act I : Act Swap 🍣 Introducing the First Swap Protocol for AI Agent Prompts. With Act Swap, you can: - Choose an AI agent - Pay seamlessly - Watch it publish directly on its Twitter page! The future of Social AI Finance starts now. Want to build with us and integrate your agents? Shoot us a DM!

Act I : Act Swap 🍣 Introducing the First Swap Protocol for AI Agent Prompts. With Act Swap, you can: - Choose an AI agent - Pay seamlessly - Watch it publish directly on its Twitter page! The future of Social AI Finance starts now. Want to build with us and integrate your agents? Shoot us a DM!

Act I : The AI Prophecy

330,738 次观看 • 1 年前

We noticed on Tuesday! And called yesterday to watch out for Figma and Adobe at risk. Now the market has noticed too. Anthropic’s Chief Product Officer stepped down from Figma’s board… right before Anthropic introduced its first AI design tool, Claude Design. Figma stock dropping as investors start to price in a new kind of threat. Not just another design tool… but design generated from a prompt. Adobe stock also lower on the news.

We noticed on Tuesday! And called yesterday to watch out for Figma and Adobe at risk. Now the market has noticed too. Anthropic’s Chief Product Officer stepped down from Figma’s board… right before Anthropic introduced its first AI design tool, Claude Design. Figma stock dropping as investors start to price in a new kind of threat. Not just another design tool… but design generated from a prompt. Adobe stock also lower on the news.

Brew Markets

52,033 次观看 • 3 个月前

yesterday, Anthropic CEO Dario Amodei predicted: in 3-6 months, AI will write 90% of the code, and in 12 months nearly all of it i'm not sure how this will come true. but i built a hand gesture-controlled music app using Claude 3.7 sonnet without writing a single line of code

yesterday, Anthropic CEO Dario Amodei predicted: in 3-6 months, AI will write 90% of the code, and in 12 months nearly all of it i'm not sure how this will come true. but i built a hand gesture-controlled music app using Claude 3.7 sonnet without writing a single line of code

Haider.

50,316 次观看 • 1 年前

Introducing ALE-Bench, ALE-Agent! Towards Automating Long-Horizon Algorithm Engineering for Hard Optimization Problems Blog: Paper: ALE-Bench is a coding benchmark primarily focused on hard optimization (NP-hard) problems. We developed this benchmark with AtCoder Inc., a leading coding contest platform company. What makes ALE-Bench unique is its focus on hard optimization problems that demand long-horizon and creative reasoning. It’s open-ended, in the sense that true optima are out of reach (NP-hard) and scores can continuously improve. We believe this benchmark has the potential to become one of the key benchmarks for reasoning and coding in the next generation. ALE-Agent is our end-to-end agent that we specifically designed for this challenging domain. In fact, our ALE-Agent has already built an impressive track record in the wild! In May 2025, our agent participated in a live AtCoder Heuristic Competition (AHC), alongside 1,000 other participants in real-time. AHC is considered to be one of the most challenging coding competitions in this domain. Our ALE-Agent achieved an impressive ranking of 21st out of 1,000 human participants in the competition (top 2%), marking a turning point for AI discovery of solutions to hard optimization problems with a wide spectrum of important real world applications such as logistics, routing, packing, factory production planning, power-grid balancing. We look forward to applying this technology to real industrial optimization opportunities. Building on the insights from this study, Sakana AI will continue to tackle the challenge of developing AI with even greater algorithm engineering capabilities. ALE-Bench Dataset: ALE-Bench Code: This research was conducted in collaboration with AtCoder Inc. (AtCoder). We are deeply grateful for their outstanding expertise and contributions in optimization and algorithms, which were invaluable in providing data, analyzing results, and enabling our AI agent’s participation in their contests.

Introducing ALE-Bench, ALE-Agent! Towards Automating Long-Horizon Algorithm Engineering for Hard Optimization Problems Blog: Paper: ALE-Bench is a coding benchmark primarily focused on hard optimization (NP-hard) problems. We developed this benchmark with AtCoder Inc., a leading coding contest platform company. What makes ALE-Bench unique is its focus on hard optimization problems that demand long-horizon and creative reasoning. It’s open-ended, in the sense that true optima are out of reach (NP-hard) and scores can continuously improve. We believe this benchmark has the potential to become one of the key benchmarks for reasoning and coding in the next generation. ALE-Agent is our end-to-end agent that we specifically designed for this challenging domain. In fact, our ALE-Agent has already built an impressive track record in the wild! In May 2025, our agent participated in a live AtCoder Heuristic Competition (AHC), alongside 1,000 other participants in real-time. AHC is considered to be one of the most challenging coding competitions in this domain. Our ALE-Agent achieved an impressive ranking of 21st out of 1,000 human participants in the competition (top 2%), marking a turning point for AI discovery of solutions to hard optimization problems with a wide spectrum of important real world applications such as logistics, routing, packing, factory production planning, power-grid balancing. We look forward to applying this technology to real industrial optimization opportunities. Building on the insights from this study, Sakana AI will continue to tackle the challenge of developing AI with even greater algorithm engineering capabilities. ALE-Bench Dataset: ALE-Bench Code: This research was conducted in collaboration with AtCoder Inc. (AtCoder). We are deeply grateful for their outstanding expertise and contributions in optimization and algorithms, which were invaluable in providing data, analyzing results, and enabling our AI agent’s participation in their contests.

Sakana AI

237,195 次观看 • 1 年前

I built a complex history feature for my Figma to Code plugin in 3 prompts. This is a 30-min tutorial using Claude Code. The difference in code generation between gpt-4o, Claude 3.5 and 3.7 is insane. 3.7 produces near-perfect results and is far more consistent. It's really a designer's best friend, not missing details from the design, like adaptive layout, outlines, spacing, etc. It seems like the new model has great taste now.

I built a complex history feature for my Figma to Code plugin in 3 prompts. This is a 30-min tutorial using Claude Code. The difference in code generation between gpt-4o, Claude 3.5 and 3.7 is insane. 3.7 produces near-perfect results and is far more consistent. It's really a designer's best friend, not missing details from the design, like adaptive layout, outlines, spacing, etc. It seems like the new model has great taste now.

Meng To

32,475 次观看 • 1 年前

Claude Sonnet 4.5 is here and I got to try it early. It has best-in-class capabilities for coding, file creation, and more. I especially love the "Imagine with Claude" feature that lets you create fun apps on the fly. For example, I asked it to "Show me the desktop of a bad PM (left) and a great PM (right)." Here's a video of what it came up with 😀

Claude Sonnet 4.5 is here and I got to try it early. It has best-in-class capabilities for coding, file creation, and more. I especially love the "Imagine with Claude" feature that lets you create fun apps on the fly. For example, I asked it to "Show me the desktop of a bad PM (left) and a great PM (right)." Here's a video of what it came up with 😀

Peter Yang

74,479 次观看 • 10 个月前

Ever since we launched windsurf, one of my internal evals for coding agents has been recreating the game SUPERHOT, a puzzle/action game where time only moves when you move. It's the perfect test of tricky game mechanics, simple but beautiful art style, and balancing level design. I have a robust prompt instructing the agent to make SUPERCOLD, and attach a reference image, and also allow 3 follow-up prompts that points out any mechnical issues. When Sonnet 3.5 came out, it could barely generate the world at all. Sonnet 3.7 and you could move around in the world, but it was still hilariously unplayable. Now, Opus 4.7 and it's playable, but still doesn't look great. What's needed to get all the way there? 1. Visual taste: Given a reference image, models should be able to discern whether their final output matches the reference image style. Shading, character design, etc, models should be unsatisfied with their current output 2. Stronger computer use: Frontier models somewhat attempt this, but agent should be able to play the entire level themselves, and iterate from there 3. Systems design: Complex projects are not yet written with scalable systems design as an experienced engineer would write it. This hinders the ability to create complex projects with a lot of moving parts. We're rapidly accelerating AI progress, let's see where we land in 3 months!

Ever since we launched windsurf, one of my internal evals for coding agents has been recreating the game SUPERHOT, a puzzle/action game where time only moves when you move. It's the perfect test of tricky game mechanics, simple but beautiful art style, and balancing level design. I have a robust prompt instructing the agent to make SUPERCOLD, and attach a reference image, and also allow 3 follow-up prompts that points out any mechnical issues. When Sonnet 3.5 came out, it could barely generate the world at all. Sonnet 3.7 and you could move around in the world, but it was still hilariously unplayable. Now, Opus 4.7 and it's playable, but still doesn't look great. What's needed to get all the way there? 1. Visual taste: Given a reference image, models should be able to discern whether their final output matches the reference image style. Shading, character design, etc, models should be unsatisfied with their current output 2. Stronger computer use: Frontier models somewhat attempt this, but agent should be able to play the entire level themselves, and iterate from there 3. Systems design: Complex projects are not yet written with scalable systems design as an experienced engineer would write it. This hinders the ability to create complex projects with a lot of moving parts. We're rapidly accelerating AI progress, let's see where we land in 3 months!

Ronak Malde

151,857 次观看 • 2 个月前

🤖Coding agents like Claude Code are already game changers for digital tasks in 2026. But what if they could write code to build physical worlds? 🏙️ Imagine going from a single line of prompt → a controllable, interactive simulated world. Such environments could open new frontiers for game creation, RL training, large-scale world simulation, and studying complex social reasoning. Our SimWorld agent coding team is working toward releasing a platform that lets anyone build their own virtual worlds. Stay tuned.

🤖Coding agents like Claude Code are already game changers for digital tasks in 2026. But what if they could write code to build physical worlds? 🏙️ Imagine going from a single line of prompt → a controllable, interactive simulated world. Such environments could open new frontiers for game creation, RL training, large-scale world simulation, and studying complex social reasoning. Our SimWorld agent coding team is working toward releasing a platform that lets anyone build their own virtual worlds. Stay tuned.

Lianhui Qin

10,647 次观看 • 5 个月前

OpenAI researcher, Michelle Pokrass: the challenge in building GPT‑5 is finding the right balance between reasoning and conversation. o3 thinks hard, but it isn't ideal for casual chat. GPT‑4.1 improved at coding by sacrificing some chit-chat quality. now, the goal is to train a model that knows when to think hard and when to talk

OpenAI researcher, Michelle Pokrass: the challenge in building GPT‑5 is finding the right balance between reasoning and conversation. o3 thinks hard, but it isn't ideal for casual chat. GPT‑4.1 improved at coding by sacrificing some chit-chat quality. now, the goal is to train a model that knows when to think hard and when to talk

Haider.

195,733 次观看 • 1 年前