ylsCode
This service is powered by ELISE AGI. For more services, visit the ELISE AGI website.
Choose the service package that fits your needs. See the Model Pricing page for supported models. More models are being added.
Subscription packages are suitable for users who need continuous server usage. You pay by subscription cycle, and long-term use provides better value. Codex subscriptions are available for the latest GPT and other supported models.
Pay-as-you-go billing charges only for actual usage. It provides more model choices and flexible usage. Top up your account as needed, and the system automatically deducts the corresponding fee when you use the API. See the model pricing page for details.
First claim a trial/subscription plan, then get your API KEY from the account center.
Choose CLI, plugin, or APP based on your needs. CLI supports more new features, such as multi-agent concurrency and remote context compaction.
Plugins are suitable for users who are less familiar with command lines. The APP is suitable for users who prefer a graphical interface and can manage multiple projects at the same time.
high is suitable for most scenarios and balances speed and quality. Reasoning level xhigh is suitable for more complex reasoning scenarios, but it is slower and not ideal for simple tasks.name = "OpenAI" in config.toml to enable automatic compaction for long contexts, or use /compact to compact manually./model - Switch models. Choose the latest model for better performance and response quality.
/new - Start a new session, clear history, and begin a new conversation.
/plan - Planning mode. Discuss before applying changes. This is useful for planning changes, project refactoring, and scenarios that should be executed according to a plan.
/fast - Fast mode. It consumes more token resources but responds faster, with an estimated 30% speed improvement. Use it when response speed matters.
/compact - Manually compact context. Long contexts can also trigger remote context compaction automatically.