
OpenCode
Work on a repository with a terminal coding agent and your choice of model
Gallery
4 items




About OpenCode
OpenCode’s provider configuration gives a team control over where model requests go. For example, a developer can use a supported hosted endpoint for one repository and a local Ollama runtime for another. That choice also determines the available context window, inference cost and hardware requirements. For a rollout, evaluate the model and the client together. A representative repository task should include finding the relevant files, proposing an edit and running the project’s checks. Configure permission rules for file and command access before comparing the results. This assessment is based on the published product documentation.
OpenCode gives developers a terminal interface for working with an AI agent inside a codebase. A maintainer can ask the agent to trace a feature across files, propose a change and edit the implementation within configured permissions. This suits teams that want to keep their terminal workflow while choosing which model receives project context.
Provider configuration is a central part of deployment. Developers can connect supported hosted services, configure custom endpoints or use a local runtime such as Ollama. Switching providers changes the available context window, latency and inference costs. Teams can set rules governing tool actions and directory access, which helps distinguish routine repository reads from edits or commands that need review.
Key Features
- Terminal-based agent sessions
- Hosted model providers and custom endpoints
- Local model connections through Ollama
- Configurable tool permissions and directory rules
- MCP server integration for additional tools
Pros
- Use the same development workflow with different model providers
- Integrate a local model into a repository workflow
- Define permission rules for agent actions
Cons
- Hosted model charges depend on provider usage
- Local inference needs sufficient hardware for the chosen model
- Provider setup and context limits can affect large repository tasks
Pricing & Plans
The open-source client is free to install. Model inference is priced separately by the selected provider or subscription. Using a local runtime incurs hardware and operating costs.
Claim Verified Creator Badge
Are you the founder of OpenCode? Display this listing's verified badge on your website to show your customers that your product has been vetted and listed on AI Central Resources.
<a href="https://www.aicentralresources.com/tool/opencode" target="_blank" rel="noopener"> <img src="https://www.aicentralresources.com/badges/featured-badge-dark.svg" alt="Featured on AICentralResources" width="200" height="54" style="border: none;" /> </a>
* Place this HTML snippet in your website's footer, landing page, or press section. This creates a search-friendly backlink directly to your verification page.
Frequently Asked Questions
Hosted inference depends on the chosen provider and account. An installed client can also connect to a local model runtime.
OpenCode documents permission rules for tool actions and directories. Teams can configure approvals for actions that require review.
Similar AI Tools to OpenCode
View all alternatives of OpenCode
Amp
Multi-model AI coding agent — free tier, no token caps, Librarian codebase search

Warp
The open platform for automating development — terminal, agent CLI, and cloud factories




