impulse

Text Embedding 004

Frontier labEmbeddingRetiring 1 Apr 2027
API id text-embedding-004
A newer model in this line is out: Text Embedding 005.
Vertex AI retires it on 1 Apr 2027 (in 133 days). No API serves it after that.
Input / 1M
$0.10
at Vertex AI
Output / 1M
$0.00
at Vertex AI
Context window
2K
tokens
Max output
-
tokens per reply
Hosts
1
1 listings
Released
-
date not published

Pricing

List prices at Vertex AI, in US dollars per million tokens.

TierInput / 1MOutput / 1M
Standard$0.10$0.00

What it costs a month

At Vertex AI list prices, 30 days.

  • Chat assistant5,000 requests a day, 2K in / 0.5K out$30.00
  • RAG over documents2,000 requests a day, 8K in / 0.6K out$48.00
  • Coding agent1,000 requests a day, 40K in / 2K out$120
  • Overnight batch20,000 requests a day, 4K in / 0.4K out$240
Use your own numbers

Where to run it

1 host serve Text Embedding 004. The comparison is on blended price, 3 parts input to 1 part output.

HostIn / out per 1M
text-embedding-004
$0.10 / $0.00
Retires 1 Apr 2027

Capabilities

Core

  • Streaming
  • Function calling
  • JSON mode
  • Structured outputs

Multimodal

  • Vision
  • Image input
  • Audio
  • Image generation
  • Video

Advanced

  • Reasoning
  • Tool calling
  • MCP
  • Prompt caching
  • Batch API
  • Embeddings
Takes in
Text
Produces
Embeddings

Cheaper alternatives

Current models that cost less and keep most of what this one can do: tools, vision, reasoning, structured output, caching and enough context.

Same model, cheaper host

No host sells it for less than Vertex AI.

Smaller Google models

No cheaper Google model keeps its key capabilities.

Other frontier labs

Open-weight picks

Model line

Newer and older versions, at each lab's own price.

  1. UndatedText Embedding 005Latest$0.10 / $0.00
  2. UndatedText Embedding 004This page$0.10 / $0.00

Call it

Using Vertex AI's API.

Python
from google import genai

client = genai.Client()  # reads GEMINI_API_KEY

response = client.models.generate_content(model="text-embedding-004", contents="Hello")
print(response.text)

Quick answers

How much does Text Embedding 004 cost?

On Vertex AI, Text Embedding 004 costs $0.10 per million input tokens.

Which APIs offer Text Embedding 004?

Text Embedding 004 runs on 1 API: Vertex AI.

What is Text Embedding 004's context window?

2K tokens.

Is Text Embedding 004 being retired?

Vertex AI retires it on 1 Apr 2027. No API serves it after that.

Is there a newer version of Text Embedding 004?

Yes. Text Embedding 005 is the newest model in this line.

What are cheaper alternatives to Text Embedding 004?

Models that cost less and keep most of what it can do: Text Embedding 3 Small on OpenAI (80% less) and Snowflake Arctic Embed L V2.0 on Snowflake (30% less).

What is Text Embedding 004's API model id?

Use text-embedding-004 with Vertex AI.

Sources

Raw catalog entry for text-embedding-004 on Vertex AI
{
  "deprecation_date": "2027-04-01",
  "deprecation_source": "https://cloud.google.com/vertex-ai/generative-ai/docs/learn/model-versions",
  "input_cost_per_character": 2.5e-8,
  "input_cost_per_token": 1e-7,
  "litellm_provider": "vertex_ai-embedding-models",
  "max_input_tokens": 2048,
  "mode": "embedding",
  "output_cost_per_token": 0,
  "output_vector_size": 768,
  "source": "https://cloud.google.com/vertex-ai/generative-ai/docs/learn/models"
}