Google Launches Gemini Computer Googlebook, Embedding AI into the Cursor
Google launched its first Googlebook laptop tailored for Gemini at the Android Show, featuring a smart cursor and supporting cross-device experiences. It also released Android Gemini Intelligence, supporting cross-app automatic multi-step task execution, Chrome-integrated Gemini plugins, Rambler voice-to-text, and custom desktop widgets. Google partnered with Apple to add end-to-end encrypted RCS and previewed smart glasses at next week's I/O conference.
Gemini Official Website Entry: Google AI Gemini 3.1 Domestic Usage Guide
Update Time: 2026-07-06
Gemini (Gemini) is the flagship AI model developed by the Google DeepMind team, renowned for its excellent multimodal capabilities and deep logical reasoning. It can process text, images, audio, and video, and generate high-quality code, widely regarded as a top representative of current AI technology.
In February 2026, Google released Gemini 3.1 Pro, achieving a score of 77.1% in the ARC-AGI-2 benchmark test, more than double that of the previous Gemini 3 Pro, comprehensively surpassing GPT-5.2 and Claude Opus 4.6. In April 2026, Google also launched Gemini 3.1 Flash TTS, supporting text-to-speech control in 70+ languages, with 200+ built-in audio tags for fine-grained control of speech speed, tone, and emotion.
This article details how to use Gemini domestically and its latest model capabilities.
Gemini Domestic Availability Solutions
For domestic users, direct access to Gemini presents certain obstacles. The following provides several common solutions:
Using Chinese services that access the Gemini AI API
There are multiple service providers globally that provide Chinese access interfaces based on the Gemini API, and users can choose according to their needs. It is important to note the following points when choosing:
- Confirm whether the service provider transparently states the use of the official API
- Understand the data privacy policy
- Prioritize platforms with clear operating entities and customer service channels
For example, platforms such as the Lazyman.ai aggregation platform and huoya.ai mirror provide aggregated access entries, helping users use Gemini-related capabilities in the domestic network environment.
Official Channels
- Gemini Official Website Address: https://gemini.google.com
- Google AI Studio Official Homepage: https://aistudio.google.com
Please recognize the above as the only channels for the ChatGPT official website
What is Gemini?
Gemini is a multimodal AI system created by Google DeepMind, with the current latest series being Gemini 3.x. Core members include:
Core Capabilities of Gemini 3.1 Pro
Top-tier Reasoning Performance: Gemini 3.1 Pro scored 77.1% in the ARC-AGI-2 benchmark test (more than double that of Gemini 3 Pro), reached 94.3% in the GPQA Diamond test, and 44.4% in Humanity's Last Exam, comprehensively leading GPT-5.2 and Claude Opus 4.6.
Ultra-long Context: Supports up to 1 million input tokens and 64,000 output tokens, capable of processing entire books, codebases, or long documents at once.
Code Generation and Animation: Gemini 3.1 Pro can directly generate website-level animated SVGs from text prompts (remaining clear at any scale), build real-time space dashboards (connecting to International Space Station telemetry data), and write complex 3D interactive experiences.
Creative Programming: Can transform literary themes into functional code. For example, after reading the atmospheric tone of "Wuthering Heights," it can design a modern literary work display website.
Core Capabilities of Gemini 3.1 Flash TTS
Fine-grained Voice Control: Directly control speech speed, tone, emotion, and style through 200+ audio tags (such as [whispered], [excited], [shouting]).
Multilingual and Accents: Supports 70+ languages, offering rich regional accent choices, including American, British, Southern accents, as well as local accent variants like Brixton and RP.
Scene Templates: Built-in scene templates such as podcast dialogue, audiobook narrator, language tutor, voice assistant, health guide, news broadcast, and customer service representative, which can be called directly.
Multi-character Dialogue: Natively supports multi-character dialogue, where each character can maintain consistent voice and natural turn-taking rhythm, suitable for podcast and conversational agent development.
Ranking Performance: Ranked second on the Artificial Analysis TTS leaderboard with 1211 points, leading in quality-to-cost ratio.
How to Use Gemini Domestically?
Access via Google Official
Gemini officially provides the following products; domestic users need to resolve network access themselves:
- Gemini App: https://gemini.google.com — Consumer-grade web chat interface
- Google AI Studio: https://aistudio.google.com — Developer API debugging platform
- Vertex AI: Enterprise-level deployment solution (Google Cloud)
- Gemini CLI: https://geminicli.com — Command-line tool
Domestic Direct Solutions
If unable to access Google services, Gemini can be used via the following domestic platforms:
- gemini-cnblog — Supports Gemini 3 Pro, Gemini 3.1 Pro, and multiple models such as Claude and Grok
- gemini-docs.com — Supports Gemini 3.1 Pro and various AI models, scan code to use
- lazymanchat.com — Gemini Chinese version entry
Gemini 3.1 Pro Interface
Gemini 3.1 Pro Technical Specifications Overview
Frequently Asked Questions (FAQ)
Q: What improvements does Gemini 3.1 Pro have compared to 3 Pro?
A: Gemini 3.1 Pro is more than double that of 3 Pro in the ARC-AGI-2 benchmark, supports longer output tokens (64,000), and has significantly enhanced capabilities in creative programming, complex system synthesis, and animation generation.
Q: What scenarios is Gemini 3.1 Flash TTS suitable for?
A: Very suitable for scenarios requiring fine-grained voice control such as audiobook production, podcast recording, multilingual voice applications, customer service dialogue robots, and language learning assistants.
Q: Is it safe to use mirror sites domestically?
A: All recommended sites have been screened. However, it is recommended not to input sensitive personal information (such as real name, ID number, password, etc.) in conversations.
Q: Are there usage limits for the Gemini model?
A: Using via official channels has free quotas and rate limits. The restriction policies of domestic mirror sites vary, and some advanced features may require payment.
Q: What is Gemini CLI?
A: Gemini CLI is a command-line tool provided by Google, allowing developers to directly call the Gemini API via terminal, suitable for local development and script integration.
This article is compiled based on the latest news from April 2026. The Gemini model is continuously iterating; capabilities are subject to Google's official release.
This page provides an editorial summary based on publicly available information. It is not a republished article. Use the source link below for the original report.