ag.dev

What It Is

Ag.dev positions itself as “The Agent Development Platform,” focused on building, testing, and scaling autonomous agents without code to automate a wide range of tasks. Its core pricing concept is the agent run: one agent task from start to finish counts as a single run, whether it lasts 30 seconds or 30 minutes.

Core Capabilities and Integrations

Based on the publicly available text, Ag.dev is focused less on a single chatbot and more on autonomous agents that can run at scale. The platform supports scaling from one agent run to thousands of runs, and the Standard plan has no artificial usage limits. For models, users can bring their own OpenAI, Anthropic, or other LLM provider API keys, or have LLM usage deducted from their account balance. This is friendly to teams that want control over model providers and costs. However, the page does not disclose details about agent orchestration capabilities, tool calling, built-in templates, evaluation and debugging features, or observability.

Pricing and Trial

Ag.dev’s pricing is relatively transparent: the Standard plan costs $0.15 per agent run, plus 5% of the actual LLM usage cost. LLM costs are not included in the run fee and are always billed separately. New users receive $2 in free credits for initial exploration. Enterprise is aimed at larger customers, high-volume usage, or teams requiring custom integrations, and requires contacting sales. Overall, pay-as-you-go pricing with no monthly fee and no long-term commitment is well suited to experimentation, but for complex tasks or high-frequency runs, LLM cost will become the main variable.

Pros, Cons, and Best Fit

The advantages are simple onboarding, no-code usage, a clear pricing structure, support for user-owned API keys, plus documentation and a Discord community. Enterprise customers can get priority support, dedicated channels, and response-time guarantees. The limitations are that public information is relatively sparse: there is no clear information on Chinese-language support, data privacy, log retention, compliance capabilities, output quality, failure handling, or safety boundaries. As a result, it is better suited to developers, automation enthusiasts, and early-stage teams quickly validating agent-based automation workflows. For finance, healthcare, or highly regulated enterprises, further consultation on privacy and enterprise support terms is necessary.

China Access and Alternatives

The text does not provide information on access from mainland China, payment methods, or localization, so china_access can only be rated as unknown. If access, payment, or overseas LLM API keys are restricted, alternatives such as Dify, LangGraph, CrewAI, Relevance AI, Zapier AI Agents, and Make may be worth considering, depending on whether no-code usage, private deployment, or enterprise integrations are required.