Enterprise DNAEnterprise DNA
LModels
Google logo

Gemini Embedding 2

by Google

Embedding model for semantic search, retrieval, clustering, and ranking pipelines

Visit GoogleSubmit your build →
GE

Models

Gemini Embedding 2

Added 13 July 2026

#google#temperature

Overview

Embedding model for semantic search, retrieval, clustering, and ranking pipelines

Best for

Best for
Multimodal retrieval embeddings

How to use / API access

Call via other with model id `google/gemini-embedding-2`.

  • other · google/gemini-embedding-2

OpenRouter id: google/gemini-embedding-2

Notes

Gemini Embedding 2 is listed in the EDNA Models directory, sourced from models.dev.

Pros

  • Indexing mixed source types into an embedding workflow
  • text, image, file, audio, video input; embeddings output on the listed OpenRouter route

Cons

  • Expecting an embedding endpoint to answer a question
  • Mixing vectors from different models in one similarity index
  • Base text-token rates exclude other modalities, tools, caching, retries and any hosting-specific charges. Check actual usage before comparing cost per successful task.
Free 27-page guide

Get the free Developer’s Field Guide

A 27-page field guide to the AI coding workflow with Claude. Claude Code, MCP servers, the prompt patterns that work, and what to delegate. Free.

Enter your work email. We send it straight over, plus a few short notes worth knowing. Unsubscribe any time.

No spam. Unsubscribe any time.

← Back to ModelsSubmit your own entry →

USAGE / OPENROUTER

How much Gemini Embedding 2 is used

Tokens processed through OpenRouter each day. It shows what developers actually run, not every company's private traffic.

Tokens, last 7 days
14.5B ▲9%
Requests, last 7 days
43.5M
Tool calls, last 7 days
—
02.5B5B19 Sept: 1.6B tokens, 3.8M requests20 Sept: 1.7B tokens, 3.3M requests21 Sept: 2.4B tokens, 4.7M requests22 Sept: 3.1B tokens, 4M requests23 Sept: 3.3B tokens, 3.4M requests24 Sept: 1.9B tokens, 3.7M requests25 Sept: 1.6B tokens, 4.6M requests26 Sept: 1.8B tokens, 2.9M requests27 Sept: 1.8B tokens, 4.1M requests28 Sept: 2B tokens, 4.1M requests29 Sept: 2.4B tokens, 3.9M requests30 Sept: 1.6B tokens, 4.1M requests1 Oct: 2.8B tokens, 4.6M requests2 Oct: 1.5B tokens, 3.3M requests3 Oct: 1.2B tokens, 2.9M requests4 Oct: 1.6B tokens, 3M requests5 Oct: 1.9B tokens, 5.4M requests6 Oct: 1.3B tokens, 3M requests7 Oct: 1.6B tokens, 3.4M requests8 Oct: 1.9B tokens, 7.3M requests9 Oct: 2.7B tokens, 9.3M requests10 Oct: 2.3B tokens, 8M requests11 Oct: 1.6B tokens, 5.6M requests12 Oct: 2.8B tokens, 7.8M requests13 Oct: 2.1B tokens, 8.1M requests14 Oct: 2.5B tokens, 6.7M requests15 Oct: 1.9B tokens, 5.5M requests16 Oct: 2.4B tokens, 6.5M requests17 Oct: 1.7B tokens, 5.2M requests18 Oct: 1.1B tokens, 3.7M requests19 Sept3 Oct17 Oct18 OctSee where it ranks on the leaderboard →

MODEL FIELD GUIDE / SOURCE EVIDENCE

What you can do with Gemini Embedding 2.

Specifications reviewed; no EDNA task evaluation has been run. Sources checked 2026-10-20. Hosting routes and app features can differ.

DeveloperGoogle
Exact model identifiergoogle/gemini-embedding-2
Inputtext, image, file, audio, video
Outputembeddings
Tool calling
Not established
Structured output
Not established
Reasoning controls
Not established
Availability
available
Price basis
token
Licence verification
Check the developer licence; the model name is insufficient

Base text-token rates exclude other modalities, tools, caching, retries and any hosting-specific charges. Check actual usage before comparing cost per successful task.

Consider it for

  • Indexing mixed source types into an embedding workflow
  • Comparing cross-modal retrieval with a labelled relevance set

Watch for

  • Expecting an embedding endpoint to answer a question
  • Mixing vectors from different models in one similarity index

PRACTICAL EXERCISE / NOT RUN BY ENTERPRISE DNA

Try the model on a small, checkable task.

Embed query: safety inspection checklist. Embed three owned records: a safety checklist PDF, a lunch menu and an equipment inspection photograph. Rank cosine similarity without generating an answer.

Check the result: Check dimensions and finite numbers; inspect whether relevant records outrank the menu using the labelled set.

Compare another approach

openai text embedding 3 large ↗qwen qwen3 embedding 8b ↗
Inspect the sources and their scope
provider source ↗

Developer documentation for model family or task. OpenRouter route-specific properties remain catalogue claims; EDNA has not executed a task evaluation.

Checked 2026-10-19
catalogue source ↗

google/gemini-embedding-2: listed route capabilities, availability and prices; provider endpoints can differ.

Checked 2026-10-20
How to evaluate a model for your work →

FROM SPECIFICATIONS TO YOUR WORK

Is this the right model for your job?

Start with a realistic sample. These workflows match the model’s listed capabilities, not a verified result on your task.

Understand the capabilitiesRead the model guide ↗

Check the source

Price date: 2026-10-20. Prices are USD API token rates. Your app subscription, uploads, browsing and tools are separate.

Model source & availability ↗

Understand the evidence

Listed capabilities and third-party benchmark scores do not establish Enterprise DNA task-test results. Validate accuracy and total cost on your own sample.

Compare this model with alternatives →