# Gemini Embedding 2 - Core.Today AI API

> Google's second-generation Gemini embedding model and the successor to gemini-embedding-001. Generates vector representations for semantic search, clustering, classification, and RAG at $0.20 per million text input tokens.

- **Provider**: Google
- **Model ID**: gemini-embedding-2
- **Category**: LLM
- **Credits**: 0.4 per 1K tokens
- **Speed**: Fast
- **Quality**: High

## Model Specifications
- **Context Window**: 8K tokens
- **Max Output**: 0K tokens
- **Training Cutoff**: 2025
- **Supported Formats**: text
- **Compatible SDK**: OpenAI, Google AI

### Token Pricing (per 1M tokens)
- **Input Tokens**: 371.6 credits ($0.25)
- **Output Tokens**: 0 credits ($0.00)

## Features
- Successor to gemini-embedding-001 (retiring 2028-05-14)
- Matryoshka Representation Learning (MRL) for flexible dimensionality
- Optimized for semantic search, clustering, classification, and RAG
- Text input billed at $0.20 per 1M tokens; no output-token charge

## Use Cases
- Semantic search
- Document clustering
- Similarity matching
- Recommendation systems
- RAG (Retrieval-Augmented Generation)

## API Endpoint

Base URL: https://api.core.today
Endpoint: POST /llm/gemini/v1beta/openai/embeddings

## Authentication
Header: Authorization: Bearer YOUR_API_KEY

Note: LLM endpoints use OpenAI-compatible format with Authorization Bearer token.

## Input Parameters

### Required
- **input**: string | array - Text or array of texts to embed.
- **model**: string (default: gemini-embedding-2) - Model identifier.

### Optional
- **dimensions**: integer (default: 3072) - Output embedding dimensionality (MRL). Smaller values reduce storage and search cost; L2-normalize truncated vectors before cosine similarity.

## Examples

### Text Embedding
Generate embeddings for semantic search
```json
{
  "model": "gemini-embedding-2",
  "input": "What is the meaning of life?"
}
```

## Response Format
```json
{
  "id": "chatcmpl-abc123",
  "object": "chat.completion",
  "choices": [
    {
      "index": 0,
      "message": {
        "role": "assistant",
        "content": "Response text here"
      },
      "finish_reason": "stop"
    }
  ],
  "usage": {
    "prompt_tokens": 100,
    "completion_tokens": 50,
    "total_tokens": 150
  }
}
```

## Tips
- Text input is billed at $0.20 per 1M tokens (371.6 credits/M); audio input is billed at $6.50 per 1M tokens. The OpenAI-compatible embeddings endpoint takes text input
- Vectors are not compatible with gemini-embedding-001 - re-embed your whole corpus when you switch
- Use a smaller `dimensions` value to cut vector storage and search cost
- Batch multiple texts in a single request for efficiency

## Documentation
https://ai.google.dev/gemini-api/docs/embeddings