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.

all-mpnet-base-v2 vs GTE-Base: API price, context & features · api14

Sentence Transformers: all-mpnet-base-v2 vs Thenlper: GTE-Base

A side-by-side comparison of Sentence Transformers: all-mpnet-base-v2 ($0.02 / 1M tokens) and Thenlper: GTE-Base ($0.02 / 1M tokens) using current USD prices and capabilities from the api14 catalog. Both run through the same OpenAI-compatible API.

Specifications

	Sentence Transformers: all-mpnet-base-v2	Thenlper: GTE-Base
Provider	Sentence Transformers	Thenlper
Input	$0.02 / 1M tokens	$0.02 / 1M tokens
Context window	512	512
Max output	460	460
Input	text	text
Output	embeddings	embeddings
Tool calling
Structured output
Reasoning control
Added	Nov 17, 2025	Nov 18, 2025

Key differences

Both have about the same input price.

In depth

Sentence Transformers: all-mpnet-base-v2 is an embedding model from Sentence Transformers, and Thenlper: GTE-Base is an embedding model from Thenlper. Both sit in the embeddings category, meaning they take text in and give vectors out rather than generated text. They are close siblings in purpose, released a day apart in November 2025, and they are aimed at the same job: turning a piece of text into a numeric representation you can store, compare or search with.

On the specs that matter for planning, the two are effectively identical. Both accept text only and both return embeddings only, so neither one handles images, audio or any other modality. Both have a context window of 512 tokens, which is the same ceiling, so there is no model here that lets you embed longer passages in a single call. Neither lists any supported parameters, so you are not choosing between one that exposes extra knobs and one that does not. The qualitative comparison marks input pricing as a tie, and the other comparison fields are either not meaningfully different or unknown, so nothing in the supplied facts separates them on capability, context or cost tier.

That leaves the choice resting on your own evaluation rather than on a feature checklist. Because the two models are so close in shape, the sensible approach is to pick whichever one performs better on your actual data. If you are building semantic search, deduplication, clustering or retrieval over short chunks of text, either model can serve as the encoder, and the deciding factor will be how well its vectors separate your particular queries and documents. Run a sample of your real pairs through both, compare retrieval quality or similarity ordering, and let that decide. If you have no strong signal either way, the tie on input pricing means there is no cost argument pushing you toward one over the other.

Calling either model through api14 is the same operation. The API is OpenAI-compatible, so you point your existing client at the embeddings endpoint and pass the text you want encoded. The only thing that changes between the two is the model id: use sentence-transformers/all-mpnet-base-v2 for the Sentence Transformers model, or thenlper/gte-base for the Thenlper model. Everything else in the request stays as it is, which makes switching between them a one-line change and makes side-by-side comparison straightforward.

There is no case in the supplied facts where one model is clearly the better pick over the other. They share the same context length, the same input and output modalities, the same absence of supported parameters, and a tie on input pricing. If you later find that one handles your domain better, that finding will come from your own testing, not from a documented difference between the two.

Switch with one string

Both models use the same endpoint and request shape. Change the model value from "sentence-transformers/all-mpnet-base-v2" to "thenlper/gte-base" and prompts, streaming and tools keep working.

Sentence Transformers: all-mpnet-base-v2 details & playground →Thenlper: GTE-Base 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": "thenlper/gte-base",
  "input": [
    "The quick brown fox",
    "Lorem ipsum"
  ]
});
console.log(result.data.map((d) => d.embedding.length));

Sentence Transformers: all-mpnet-base-v2 vs Thenlper: GTE-Base

Specifications

Key differences

In depth

Switch with one string

More comparisons