GPT-5.6 is here ๐ŸŽ‰ 20% off all GPT ๐ŸŽ‰ All July ๐Ÿ”ฅLearn more
Gemini

Google: Gemini Embedding 2 Preview

Embedding
google/gemini-embedding-2-preview

Gemini Embedding 2 Preview is Google's multimodal embedding model, released on 2026-03-10. It converts text, images, audio, and video into numerical vectors whose geometry captures the semantic meaning and context of the source data, producing representations that machine learning pipelines and large models can consume directly. That makes it a foundation for retrieval-augmented generation, semantic search, clustering, classification, deduplication, and similarity ranking, with one model covering all four input modalities. Each request accepts up to 8K input tokens, and text input is billed at $0.20/M tokens. Accessible via the Gemini protocol through Ofox.

Context Window
8K
Max Output Tokens
8K
Released
2026-03-10
Capabilities
VisionAudio InputVideo Input
Available Providers
GoogleCloudVertex
Supported Protocols
Geminigemini

Providers

GoogleCloudVertex
Input Tokens
$0.2/M
Output Tokens
$0/M
Image Input
$0.45/M
Audio Input
$6.5/M
Video Input
$12/M
Protocols
Geminigemini

Code Examples

from google import genai
client = genai.Client(
api_key="YOUR_OFOX_API_KEY",
http_options={"api_version": "v1beta", "base_url": "https://api.ofox.ai/gemini"},
)
response = client.models.embed_content(
model="google/gemini-embedding-2-preview",
contents="Hello, world!",
)
print(response.embeddings[0].values[:5])

Frequently Asked Questions

Google: Gemini Embedding 2 Preview on Ofox.ai costs $0.2/M per million input tokens and $0 per million output tokens. Pay-as-you-go, no monthly fees.