TRAE

TRAE brings an AI-assisted IDE, an autonomous coding agent and a separate AI workspace into one product family. It is best for developers and engineering teams that want configurable agents and deployment integrations, while professionals can use TRAE Work for broader workplace tasks. The free tier lowers the entry cost, but task ceilings, trial conditions and indexing behavior deserve close attention.

Overview

TRAE IDE combines a conventional development environment with SOLO, an autonomous agent intended to handle end-to-end coding tasks. You can create custom agents and assign sub-agents to complex work, then let those agents reach external resources through Model Context Protocol integrations. This is a stronger fit for multi-step engineering workflows than a tool limited to inline suggestions, although simultaneous cloud-task capacity depends on the plan.

TRAE Work is positioned as an AI workspace for professionals, whereas TRAE IDE is aimed at software developers and engineering teams. They address different jobs: choosing TRAE for coding does not make the Work assistant a substitute for an IDE.

Key features

IDE and autonomous workflow

The IDE provides multiple model choices, codebase indexing, a built-in terminal and extension support. SOLO can take on broader implementation work, and custom agents let teams shape repeatable processes around their own tasks. That combination suits developers who want AI participation beyond autocomplete; users seeking only lightweight completion may not need the additional agent layer.

Project integrations

TRAE SOLO integrates with Vercel for deployment and Vercel AI Gateway model access. TRAE IDE and SOLO also connect with Supabase services and Supabase MCP for backend and project operations. These integrations are most useful when a project already relies on those platforms, because they bring deployment and backend actions closer to the coding workflow.

Security and indexing

Codebase files are stored locally, but files can be temporarily uploaded for embedding during indexing and plaintext is deleted afterward. That process may matter to organizations with strict data-handling rules. Enterprise adds encryption in transit, no server-side code storage, no model-training use of code and SOC 2 Type 2 certification. Enterprise repository indexing supports up to 100,000 files and 150 million lines of code.

Pricing

TRAE uses a freemium model. The Free plan costs nothing and includes 5,000 autocomplete requests per month plus two concurrent cloud tasks. It is a practical starting point for individual developers, but the request allowance and concurrency limit can constrain heavier use.

Lite costs $3 a month, billed per month, and includes $5 of basic usage plus bonus usage and premium model availability. It is the low-cost paid option for users who want premium models without moving to the larger task tiers.

Pro has custom pricing, with $20 of basic usage, unlimited autocomplete and 10 concurrent cloud tasks. Pro+ has custom pricing with $60 of basic usage and 15 concurrent cloud tasks, while Ultra has custom pricing with $200 of basic usage and 20 concurrent cloud tasks. The higher tiers target sustained agent workloads; their advantage is capacity, not a different core IDE. A 14-day Pro trial is available only to new users who add a payment method within 48 hours of signing up, so it is not a general trial for every account.

Platforms

TRAE supports iOS, Linux, macOS, the web and Windows. Desktop downloads include macOS and Windows applications, Linux.deb packages and an iOS download. The product’s supported-platform field also identifies Windows for the IDE workflow.

Who it's for

Choose TRAE if you are a developer or engineering team that wants an IDE, autonomous coding, configurable agents, indexing and Vercel or Supabase connections in one workflow. TRAE Work broadens the offering for professionals who need an AI workspace. It is less compelling for a user who only wants simple completion, or for a security-sensitive team unwilling to permit temporary embedding uploads during indexing.

Pros and cons

Pros

Agent depth: SOLO, custom agents and sub-agents support end-to-end and complex tasks.

Useful integrations: Vercel and Supabase connections link coding with deployment and backend operations.

Broad access: A free plan and support across desktop, web and iOS make evaluation straightforward.

Enterprise controls: Encryption, no server-side code storage, no training use and SOC 2 Type 2 address organizational requirements.

Cons

Free-tier ceilings: 5,000 monthly autocomplete requests and two concurrent cloud tasks may be restrictive for frequent use.

Trial qualification: The 14-day Pro trial requires a payment method within 48 hours of sign-up.

Indexing exposure: Files may be temporarily uploaded for embedding, which can conflict with stricter policies.

Paid-price uncertainty: Pro, Pro+ and Ultra use custom pricing, making direct budget comparison difficult.

Alternatives

For a broader selection, see AI Code Editors, Integrated Development Environments and AI Code Generators.

Replit is a better match if you want Android access and a free plan with one live project and one background task.

Qoder suits users who want basic models, limited messages and BYOK on its free plan.

Tesseract is an option for users who prefer an AI-included allowance with refills billed separately.

Zed fits readers seeking a free code editor with 2,000 accepted edit predictions and external-agent or API-key support.

Kiro offers a free monthly plan with 50 credits and access to open-weight models and Claude Sonnet 4.5.

Zubvo provides a hobby plan with 100 credits per month, open-source models and up to 2,000 tab completions.

Devgrex is an alternative for local completions, agent chat and offline mode.

Google Antigravity IDE offers individuals unlimited tab completions and command requests subject to basic weekly rate limits.

Verdict

TRAE is a strong choice for developers and engineering teams that want autonomous coding, customizable agents and Vercel or Supabase workflows alongside a conventional IDE. Its main reasons to look elsewhere are the free-tier ceilings, custom-priced upper plans and temporary embedding uploads during indexing; teams that can accept those trade-offs get the most from its agent-centered workflow.