Gemini CLI

Gemini API CLI is an experimental command-line interface for the Gemini Interactions API and managed-agent platform. It can run model interactions and managed agents, generate or analyze text and media, transcribe audio or video, manage files, models, agents, environments, triggers, and webhooks, and stream results in formats including JSON, YAML, and TOON. Commands expose machine-readable usage information and request schemas, while --dry-run validates inputs and previews redacted HTTP requests without contacting the API. Authentication can use API keys, OAuth access tokens, an operating-system keychain, or a configuration file. The repository describes the CLI as beta software with possible breaking changes, generated through Speakeasy; its software is licensed under Apache 2.0, and the repository states that it is not an official Google product.

Google's developer API for building applications with Gemini large language models. It provides access to models like Gemini 2.0 Flash and 2.5 Pro, and also supports Gemma open models. Developers use it for tasks such as custom model testing, retrieval, memory, and tool-calling systems.

Gemini API managed agents are configurable agent harnesses from Google that let an application provision a Linux sandbox with a single API call. The agent can reason, execute code, manage files, and browse the web autonomously; developers can customize it with instructions, skills, data, and external tools, and package reusable agents through the Agents API. Each environment is isolated at the operating-system level and supports persistent state, although inactive environments are eventually deleted and virtual machines can spin down. Managed agents are in public preview and use pay-as-you-go pricing based on model tokens and tool usage.

Gemini is Google's AI model and assistant accessible through gemini.google.com. It processes mixed media including video, audio, images, and text, and can convert recordings into SOPs, summarize meetings, extract decisions, and draft follow-up emails. In research and development contexts, it serves as an alternative LLM backend for tasks such as adversarial audits of requirements and resume evaluation.

The Gemini Interactions API is Google's unified API for calling Gemini models and specialized agents. It maintains server-side interaction state through interaction IDs and the `previous_interaction_id` parameter, while also supporting stateless requests. Each Interaction resource contains a chronological sequence of execution steps, including model thoughts, tool calls and results, and final output. The API supports text and multimodal generation, structured and strongly typed outputs, built-in and custom tools, tool orchestration, observable execution steps, background execution for long-running tasks, and chained image, video, and audio generation using shared context. Google documents it as generally available and recommends it for new Gemini API projects.