We use cookies. Cookies and browser storage keep you signed in and remember your preferences. With your permission we also measure how the site is used. You can any time. See our cookie policy.

New here? See how it works A 1-minute tour of the playground, models and your API key.

Gemini Embedding 2 vs Embed V1 4B: API price, context & features · api14

Google: Gemini Embedding 2 vs Perplexity: Embed V1 4B

A side-by-side comparison of Google: Gemini Embedding 2 ($0.70 / 1M tokens) and Perplexity: Embed V1 4B ($0.11 / 1M tokens) using current USD prices and capabilities from the api14 catalog. Both run through the same OpenAI-compatible API.

Specifications

	Google: Gemini Embedding 2	Perplexity: Embed V1 4B
Provider	Google	Perplexity
Input	$0.70 / 1M tokens	$0.11 / 1M tokens
Audio input	$22.75 / 1M tokens	-
Image	$0.000002 / image	-
Context window	8,192	32,000
Max output	7,372	28,800
Input	text, image, file, audio, video	text
Output	embeddings	embeddings
Tool calling
Structured output
Reasoning control
Added	May 20, 2026	Mar 16, 2026

Key differences

Perplexity: Embed V1 4B is 85% cheaper on input tokens.
Perplexity: Embed V1 4B reads 3.9× more context (32,000 tokens).

In depth

Google: Gemini Embedding 2 is an embedding model from Google, listed under the model id google/gemini-embedding-2. Perplexity: Embed V1 4B is an embedding model from Perplexity, listed under the model id perplexity/pplx-embed-v1-4b. Both sit in the embeddings category, meaning both take input and return embeddings rather than generated text, and both expose no supported parameters, so you call them with your input and read the vector back.

The clearest split is in context window and input modalities. Gemini Embedding 2 accepts text, image, file, audio and video as input, and its context length is 8192 tokens. Embed V1 4B accepts text only, but its context length is 32000 tokens, so it takes noticeably more text in a single call. Neither model differs meaningfully on tool use, structured output or reasoning, since both are embedding models and those fields are not applicable here. In short, Gemini Embedding 2 is the one to reach for when your input is not plain text, while Embed V1 4B is the one to reach for when your input is a long text document.

For practical work, the choice usually follows your data rather than your preferences. If you are embedding screenshots, scanned pages, audio clips, video frames or mixed media alongside text, Gemini Embedding 2 is the only one of the two that accepts those inputs at all. If you are embedding long text passages, documentation, transcripts or whole articles and want to fit more of each item into one call, Embed V1 4B gives you the larger context window. Embed V1 4B is also the cheaper of the two on input, so text heavy workloads at volume lean its way on cost as well.

Calling either model through api14 works the same way, because the API is OpenAI compatible. You point your existing OpenAI client at api14, send a request to the embeddings endpoint, and set the model field to google/gemini-embedding-2 or perplexity/pplx-embed-v1-4b. The request shape, the response shape and your surrounding code stay the same; only the model id string changes. That means you can build one pipeline, test both models against your own data, and switch between them by editing a single value.

There are still cases where the other model wins. If your corpus is text only and the items are long, Embed V1 4B is the better pick even though Gemini Embedding 2 handles more input types, because the extra modalities buy you nothing on plain text and the larger context window does. Conversely, if a text only job fits comfortably inside 8192 tokens and you already have Gemini Embedding 2 wired up for other media, there is no strong reason to split your stack. Match the model to the input you actually have.

Switch with one string

Both models use the same endpoint and request shape. Change the model value from "google/gemini-embedding-2" to "perplexity/pplx-embed-v1-4b" and prompts, streaming and tools keep working.

Google: Gemini Embedding 2 details & playground →Perplexity: Embed V1 4B details & playground →

import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "https://api14.com/api/v1",
  apiKey: process.env.API14_API_KEY,
});

const result = await client.embeddings.create({
  "model": "perplexity/pplx-embed-v1-4b",
  "input": [
    "The quick brown fox",
    "Lorem ipsum"
  ]
});
console.log(result.data.map((d) => d.embedding.length));

Google: Gemini Embedding 2 vs Perplexity: Embed V1 4B

Specifications

Key differences

In depth

Switch with one string

More comparisons