Skip to main content
Agents are the intelligence behind how Bolt understands your requests, solves problems, and writes the code for your app. They power the building experience by turning your instructions into real features. In Bolt, you can choose the agent that best fits your work: the v1 Agent (legacy) or Claude Agent. When using Claude Agent, you can also select the underlying model (LLM) that drives it, giving you more control over speed, capability, and cost as you build.
To learn more about large language models generally, see Introduction to LLMs.

​
Build with your preferred agent

Bolt gives you the choice of which AI agent powers your builds, so you can work in the way that suits you best. You can choose between two options:
  • Claude Agent – Our most capable builder in Bolt and the best choice for production-quality apps and larger development work. It produces more complete results, makes stronger decisions, and handles complex reasoning with fewer errors. It may take a little longer to run and can use more tokens, although the higher quality often saves time overall. Claude Agent is powered by the Claude model family from Anthropic, and you can choose the level of performance you want, including fast and efficient Haiku 4.5, the everyday Sonnet 4.5/4.6, or the powerful Opus 4.5/4.6. Plan Mode is also available with Claude Agent, which helps you develop detailed build strategies and execute them accurately.
  • v1 Agent (legacy) – Based on the original Bolt experience, this agent is more likely to produce incomplete apps and require extra fixes, so it’s not recommended for production work. It is, however, faster, uses fewer tokens per prompt, and works well for quick prototypes or testing layouts and design ideas. It uses Anthropic Claude Sonnet as its LLM.
Bolt strongly recommends using Claude Agent if you want the best results and the smoothest building experience.

​
Claude Agent vs. v1 Agent (legacy) efficiency

While v1 Agent (legacy) might seem more efficient because it uses fewer tokens at the start, this can be misleading. For any project beyond a quick mockup or prototype, you will likely save tokens overall by using Claude Agent. Its more comprehensive approach to building leads to higher quality planning, greater reliability, and fewer mistakes. This means less rework, fewer wasted calls, and more efficient use of your token budget over time.

​
Switch between Bolt agents and models

Not all models are available with the free plan. To be able to switch to any model, upgrade to a paid plan.
You can switch between agents in two places: from the Bolt homepage when starting a build, or inside a project while working. For each project, Bolt remembers the model you select, and it automatically uses that model again the next time you open the project. Note that switching between agents mid-project clears your chat history, though this is intentional to help prevent confusion between agents. Your new agent will still have full access to your project and files. Any general or ongoing instructions you want to preserve should be added to Project Knowledge.
If you are already using Claude Agent, switching between models in Claude Agent does not clear your chat history.
If you want all new projects to use a specific model, you can set a default model within your personal settings. Bolt then automatically selects that model when you start a new project.

​
Switch agents or models on the homepage

Screenshot of switching agents on homepage in Bolt.
From the Bolt homepage, use the drop-down selector at the bottom of the chatbox to choose your agent. When Claude Agent is selected, you can also choose the model that powers it:
  • Haiku 4.5
  • Sonnet 4.5 (default)
  • Sonnet 4.6
  • Opus 4.5
  • Opus 4.6

​
Switch agents or models in a project

Screenshot of switching agents in a Bolt project.
To switch agents when working inside a project, follow these steps:
  1. In the bottom-left corner of the chatbox, click the current model name.
  2. Select the agent or model you want to use.
  3. Confirm your choice.
Hover over a model’s name to see more details about it.

​
Differences between agents

Bolt behaves slightly differently depending on which agent you choose. Some tools and interface elements adjust to match each agent’s strengths. Key differences include:
  • Claude Agent works with Plan Mode. v1 Agent (legacy) works with Discussion Mode.
  • Claude Agent lets you choose between Haiku, Sonnet, and Opus models..
  • Claude Agent can create Bolt Databases and connect projects to Supabase.
FeatureClaude Agentv1 Agent (legacy)
Plan Mode✔✗
Create Bolt databases✔✗
Select your model✔✗
Works with Bolt databases✔✔
Connects to Supabase✔✔
Discussion Mode✗✔

​
Model selection inside Claude Agent

Model selection controls how Claude Agent behaves in your project. Each model reflects a different balance of speed, reasoning depth, and token usage. For most users, Sonnet 4.5 is the default and recommended starting point. It offers the most predictable, well-rounded building experience across the widest range of tasks. Haiku 4.5 fits lightweight or high-volume tasks. Sonnet 4.6 brings deeper reasoning for complex, multi-step work. Opus 4.5 and Opus 4.6 are best suited for the hardest problems, large architectural decisions, or work where accuracy matters more than speed.
These models are not a simple ladder where each one is strictly “better” than the last. Each model reflects a different set of tradeoffs between speed, reasoning depth, token cost, and behavior. A model that excels at one type of task may not be the best fit for another. The best way to find what works for your projects is to experiment. Try different models on the kinds of tasks you do most and see which one gives you the results you want.

​
Haiku 4.5

haiku-4-5_jiu8vz

Fast and token-efficient for simple tasks and rapid iteration.Best for:
  • Quick UI or styling changes
  • Content edits and translations
  • Test or seed data
  • Straightforward static pages
  • Tasks where the solution is already clear

​
Sonnet 4.5

sonnet-4-5_qi5vez

The default model in Bolt. Balanced, reliable, and well-suited to the broadest range of development work.Sonnet 4.5 is the model most users should start with, and the one Bolt selects by default. It delivers strong reasoning across frontend and backend tasks while keeping token usage efficient and response times fast. Its behavior is consistent and predictable, which makes it a dependable choice whether you are scaffolding a new project, building out features, debugging, or refactoring. For most development sessions, Sonnet 4.5 handles the work without needing you to think about which model to use.Best for:
  • Scaffolding new apps and features
  • Backend logic and database work
  • Importing Figma designs into UI
  • Routine refactoring and debugging
  • General-purpose development where reliability and speed both matter

​
Sonnet 4.6

sonnet-4-6_mivwaa

Opus-class capability at a pace and price built for all-day use.Sonnet 4.6 represents a major step up in reasoning depth within the Sonnet family. It can hold a complex plan across multiple steps and execute without losing the thread.

Multi-file refactors, end-to-end feature builds, and coordinated front-and-backend changes that previously required switching to Opus are handled smoothly. For the vast majority of complex development work, Sonnet 4.6 delivers results comparable to Opus while staying fast and token-efficient enough to use throughout the day.Best for:
  • Multi-step feature builds spanning front and backend
  • Complex refactors touching many files
  • Sustained development across larger codebases
  • Any task where you would have previously reached for Opus, but want faster, more cost-effective output

​
Opus 4.5

opus-h_hsyui1

Deep reasoning for complex, system-level decisions.Opus 4.5 provides strong reasoning depth for work that goes beyond everyday feature building. It handles large refactors, multi-tenant architectures, and compliance-sensitive workflows where mistakes are costly. While Sonnet 4.6 now covers much of the territory Opus 4.5 previously owned, Opus 4.5 remains a solid choice when you want high reliability on complex problems without needing the prompt-sensitivity of Opus 4.6.Best for:
  • Large refactors across many files
  • Enterprise or multi-tenant architectures
  • Sensitive or compliance-heavy workflows
  • Problems where mistakes are costly

​
Opus 4.6

opus-4-6_jtuju3

The most capable model available in Bolt, with the deepest reasoning and greatest sensitivity to your instructions.Opus 4.6 is tuned to respond more directly to how you prompt it. If you tell it to think carefully about a specific problem, it will invest deeper reasoning in that area in a way other models may not. This makes it especially well-suited for navigating large, existing codebases and reasoning through messy or legacy code where understanding tangled dependencies and older patterns matters. For new projects or tasks where Sonnet 4.6 or Opus 4.5 already perform well, Opus 4.6 may not show a meaningful difference. Its strength emerges in the hardest problems where precise, directed reasoning makes a real impact.Best for:
  • The most complex architectural and system-wide decisions
  • Brownfield development in large, tangled codebases
  • Tasks where you want fine control over reasoning depth through your prompts
  • High-stakes work where the extra reasoning investment is justified

​
Comparing models

ModelSpeedReasoning depthBest use
Haiku 4.5FastestLightQuick edits and iteration
Sonnet 4.5 (default)BalancedStrongMost everyday development work
Sonnet 4.6BalancedDeepComplex multi-step tasks and larger codebases
Opus 4.5SlowerDeepHigh-stakes or system-wide decisions
Opus 4.6SlowerDeepestAdvanced reasoning, legacy codebases, prompt-directed deep thinking
A common workflow is to rely on Sonnet 4.5 as your default for most development, reach for Sonnet 4.6 when tasks involve deeper multi-step reasoning or larger codebases, use Haiku 4.5 for fast follow-ups or polish, and switch to Opus 4.5 or 4.6 when you need the deepest reasoning on the hardest problems. Opus 4.6 is particularly worth reaching for when working in large existing codebases or when you want the model to respond precisely to detailed prompting about where to focus its reasoning.

​
Using claude.md files with agents

Bolt agents now support importing claude.md files directly into projects. When a claude.md file is present, it is detected and used automatically, with no additional setup required. To upload a file, follow the steps below:
  • Log in to your project.
  • Click the plus icon in the chatbox.
  • Click Attach file, and choose your file.
Using a claude.md file is especially helpful for users who rely on Project Knowledge settings. Instead of adding content manually through the UI, you manage project context in Markdown files. This makes it easier to include links, organize information across multiple files, and maintain project knowledge in a more flexible and scalable way.
Important: The claude.md file acts as the entry point for agent instructions. While it can reference other files, such as text or additional Markdown files, the primary instructions must live in claude.md itself. The file name is required and cannot be changed. Agents only look for instructions starting from claude.md, then follow any links or references you include from there.

​
How Bolt handles context

All Claude models support large context windows. However, to keep performance smooth, Bolt limits the active history to recent messages. This helps reduce token usage and keeps building responsive. If a detail from earlier becomes important again, briefly restate it to bring it back into context. For any ongoing context or instructions that are important, be sure to add them to either your Project Knowledge, Account Knowledge, or, if applicable, Teams Knowledge.