shield_lock
金大哥
← 返回 Skill 列表
extension
分类: 内容与媒体无需 API Key

gemini-api-full

Google的Gemini API综合参考。在构建以下应用程序时使用：(1) Gemini模型（Gemini 3 Pro、2.5 Flash/Pro/Flash-Lite）用于文本和多模态生成，(2) 图像生成（Imagen、Nano Banana）、视频（Veo 3.1）、音乐（Lyria），(3) 函数调用、结构化输出和代理工作流，(4) 内置工具：谷歌搜索、地图、代码执行、URL上下文、计算机使用、文件搜索，(5) 实时API支持实时语音/视频流，(6) 长上下文（100万+令牌）、嵌入、文档/音频/视频理解，(7) 批处理API、上下文缓存、安全设置。触发词: 'gemini api', 'google ai', 'genai sdk', 'gemini model', 'veo', 'imagen', 'nano banana', 'lyria', 'live api', 'vertex ai'

person作者: jakexiaohubgithub
open_in_new仓库download下载资源包

Gemini API Skill

Build AI applications with Google's Gemini models and tools.

Quick Start

Installation

# Python
pip install google-genai

# JavaScript/Node.js
npm install @google/genai

# Go
go get google.golang.org/genai

Environment Setup

export GEMINI_API_KEY="your-api-key"

Basic Usage

Python:

from google import genai

client = genai.Client()
response = client.models.generate_content(
    model="gemini-2.5-flash",
    contents="Your prompt here"
)
print(response.text)

JavaScript:

import { GoogleGenAI } from "@google/genai";

const ai = new GoogleGenAI({});
const response = await ai.models.generateContent({
    model: "gemini-2.5-flash",
    contents: "Your prompt here"
});
console.log(response.text);

REST:

curl "https://generativelanguage.googleapis.com/v1beta/models/gemini-2.5-flash:generateContent" \
  -H "x-goog-api-key: $GEMINI_API_KEY" \
  -H 'Content-Type: application/json' \
  -d '{"contents": [{"parts": [{"text": "Your prompt here"}]}]}'

Model Selection

| Model | Best For | Context Window | |-------|----------|----------------| | Gemini 3 Pro | Most intelligent tasks, multimodal reasoning, agentic | See models-overview | | Gemini 2.5 Pro | Complex reasoning, coding, extended thinking | 1M tokens | | Gemini 2.5 Flash | Balanced performance, general tasks | 1M tokens | | Gemini 2.5 Flash-Lite | High-volume, cost-sensitive, fastest | See models-overview | | Imagen | High-fidelity image generation | N/A | | Veo 3.1 | Video generation (8s, 720p/1080p with audio) | N/A | | Nano Banana | Native image gen with Gemini 2.5 Flash | N/A | | Nano Banana Pro | Native image gen with Gemini 3 Pro | N/A |

Reference Documentation Index

Getting Started

| Topic | File | Description | |-------|------|-------------| | Setup & Libraries | getting-started.md | API keys, SDK installation, OpenAI compatibility |

Models & Pricing

| Topic | File | Description | |-------|------|-------------| | Model Overview | models-overview.md | All models, capabilities, context windows | | Pricing | api-pricing.md | Token costs, tool pricing | | Rate Limits | rate-limits.md | RPM/TPM limits, quotas | | Gemini 3 Guide | gemini-3.md | Gemini 3 specific features and best practices | | Imagen | imagen.md | Image generation with Imagen model | | Embeddings | embeddings.md | Text embeddings for search/RAG | | Veo | veo.md | Video generation with Veo 3.1 (69K) | | Lyria | lyria.md | Music generation with Lyria RealTime | | Robotics | robotics.md | Gemini Robotics-ER 1.5 (42K) |

Core Capabilities

| Topic | File | Description | |-------|------|-------------| | Text Generation | text-generation.md | Text generation, system instructions (38K) | | Image Gen (Nano Banana) | image-generation-gemini.md | Native image generation with Gemini (LARGE: 174K) | | Image Understanding | image-understanding.md | Vision, image analysis | | Video Understanding | video-understanding.md | Video analysis, timestamps | | Document Understanding | document-understanding.md | PDF and document processing | | Speech Generation | speech-generation.md | Text-to-speech (TTS) | | Audio Understanding | audio-understanding.md | Audio analysis, transcription |

Advanced Features

| Topic | File | Description | |-------|------|-------------| | Thinking Mode | thinking.md | Extended reasoning capabilities | | Thought Signatures | thought-signatures.md | EDGE CASE ONLY: Manual signature handling when NOT using official SDKs | | Structured Outputs | structured-outputs.md | JSON schema responses | | Function Calling | function-calling.md | Custom tool integration (54K) | | Long Context | long-context.md | 1M+ token handling, context caching |

Tools

| Topic | File | Description | |-------|------|-------------| | Tools Overview | tools-overview.md | Built-in tools summary, agent frameworks | | Google Search | google-search.md | Web search grounding | | Google Maps | google-maps.md | Location-aware grounding | | Code Execution | code-execution.md | Python code execution tool | | URL Context | url-context.md | URL content extraction | | Computer Use | computer-use.md | Browser automation (preview) (44K) | | File Search | file-search.md | RAG with document indexing |

Live API (Real-time Streaming)

| Topic | File | Description | |-------|------|-------------| | Getting Started | live-api-getting-started.md | Low-latency voice/video interactions | | Capabilities Guide | live-api-capabilities.md | Full capabilities and configurations (32K) | | Tool Use | live-api-tools.md | Function calling & Search in Live API | | Session Management | live-api-sessions.md | Session handling, time limits | | Ephemeral Tokens | ephemeral-tokens.md | Short-lived auth for client-side WebSockets |

Guides

| Topic | File | Description | |-------|------|-------------| | Batch API | batch-api.md | Async processing at 50% cost (47K) | | Files API | files-api.md | Upload and manage media files (49K) | | Context Caching | context-caching.md | Implicit & explicit caching for cost savings | | Media Resolution | media-resolution.md | Control token allocation for media | | Tokens | tokens.md | Understand and count tokens | | Prompt Design | prompt-design.md | Prompt strategies and best practices (47K) | | Logs & Datasets | logs-datasets.md | Enable logging, view in AI Studio | | Data Logging & Sharing | data-logging-sharing.md | Storage and management of API logs | | Safety Settings | safety-settings.md | Adjust safety filters | | Safety Guidance | safety-guidance.md | Best practices for safe AI use |

Troubleshooting & Migration

| Topic | File | Description | |-------|------|-------------| | Troubleshooting | troubleshooting.md | Diagnose and resolve common API issues (25K) | | Vertex AI Comparison | vertex-ai-comparison.md | READ ONLY IF USER MENTIONS "VERTEX AI": Gemini Developer API vs Vertex AI differences |

API Reference (Technical Endpoints)

Note: These are technical endpoint specifications with schemas and parameter details. For usage guides and code examples, see the guide files above.

| Topic | File | Description | |-------|------|-------------| | Overview | api-reference-overview.md | REST/streaming/realtime API overview (33K) | | Models Endpoint | api-models-reference.md | models.get, models.list, models.predict | | Generate Content | api-generate-content.md | generateContent + all response types (LARGE: 166K) | | Live API WebSockets | api-live-websockets.md | WebSockets API for Live API (48K) | | Live Music WebSockets | api-live-music-websockets.md | WebSockets API for Lyria RealTime | | Files Endpoint | api-files-reference.md | Upload/manage media files (40K) | | Batch Endpoint | api-batch-reference.md | Batch processing endpoints (40K) | | Caching Endpoint | api-caching-reference.md | Context caching endpoints (LARGE: 89K) | | Embeddings Endpoint | api-embeddings-reference.md | Embeddings generation endpoints (30K) | | File Search Stores | api-file-search-reference.md | File Search + Documents endpoints (35K) |

Large Files - Search Patterns

For large reference files (>30K), use grep to find specific sections:

image-generation-gemini.md (174K):

grep -n "## " references/image-generation-gemini.md  # List sections
grep -n "edit" references/image-generation-gemini.md  # Find editing info
grep -n "style" references/image-generation-gemini.md  # Find style transfer

api-generate-content.md (166K):

grep -n "## " references/api-generate-content.md  # List sections
grep -n "GenerationConfig" references/api-generate-content.md  # Config options
grep -n "SafetySetting" references/api-generate-content.md  # Safety types

api-caching-reference.md (89K):

grep -n "## " references/api-caching-reference.md  # List sections
grep -n "CachedContent" references/api-caching-reference.md  # Cache types

veo.md (69K):

grep -n "## " references/veo.md  # List sections
grep -n "audio" references/veo.md  # Find audio generation info

models-overview.md (67K):

grep -n "gemini-3" references/models-overview.md
grep -n "context" references/models-overview.md

function-calling.md (54K):

grep -n "## " references/function-calling.md
grep -n "parallel" references/function-calling.md  # Parallel function calls

Common Patterns

Multimodal Input (Image + Text)

from google import genai
from google.genai import types

client = genai.Client()
response = client.models.generate_content(
    model="gemini-2.5-flash",
    contents=[
        types.Part.from_image(image_path),
        types.Part.from_text("Describe this image")
    ]
)

Function Calling

tools = [
    types.Tool(function_declarations=[{
        "name": "get_weather",
        "description": "Get weather for a location",
        "parameters": {
            "type": "object",
            "properties": {"location": {"type": "string"}},
            "required": ["location"]
        }
    }])
]

response = client.models.generate_content(
    model="gemini-2.5-flash",
    contents="What's the weather in Paris?",
    config=types.GenerateContentConfig(tools=tools)
)

Google Search Grounding

response = client.models.generate_content(
    model="gemini-2.5-flash",
    contents="What are the latest AI developments?",
    config=types.GenerateContentConfig(
        tools=[types.Tool(google_search=types.GoogleSearch())]
    )
)

Thinking Mode

response = client.models.generate_content(
    model="gemini-2.5-pro",
    contents="Solve this complex problem...",
    config=types.GenerateContentConfig(
        thinking_config=types.ThinkingConfig(thinking_budget_tokens=10000)
    )
)

Streaming

for chunk in client.models.generate_content_stream(
    model="gemini-2.5-flash",
    contents="Write a story"
):
    print(chunk.text, end="")

Key Concepts

Tool Execution Flow

Built-in tools (Google Search, Code Execution): Executed by Google

  1. Send prompt with tool config → Model executes tool → Response with grounded results

Custom tools (Function Calling): You execute

  1. Send prompt with function declarations → Model returns function call JSON
  2. You execute function, send result back → Model generates final response

Thought Signatures (Important)

  • If using official SDKs with chat feature: Thought signatures are handled automatically. No action needed.
  • If manually managing conversation history: Read thought-signatures.md for Gemini 3 Pro function calling requirements.

API Endpoints

| Endpoint | Purpose | |----------|---------| | /v1beta/models/{model}:generateContent | Standard generation | | /v1beta/models/{model}:streamGenerateContent | Streaming | | /v1beta/models/{model}:embedContent | Embeddings | | /v1beta/models/{model}:countTokens | Token counting |

Base URL: https://generativelanguage.googleapis.com