Amazon: Titan Text Embeddings V2
Titan Text Embeddings V2 is an embedding model from Amazon released on Apr 30, 2024. It reads text and writes vectors, with a 8K token context window.
- Modalities
- In / out price
- n/a
- Context
- 8K
- Max output
- 2K
- Released
- Apr 30, 2024
- Knowledge cutoff
- n/a
- Tool calling
- Structured output
- Reasoning
- Prompt caching
- Open weights
- Free to use
Supported request parameters
What you can set when you call Titan Text Embeddings V2 through an OpenAI compatible API.
- Temperature
temperature
More from Amazon
Reads text, images, audio and video, writes text and images. Reasoning.
by amazonDec 2, 2025$0.30/M input tokens$2.50/M output tokens
- Index 14.2
Reads text, images, audio and video. Reasoning.
by amazonDec 2, 20251M context$1.25/M input tokens$10/M output tokens
- 1M context
Reads text, images, files and video. Tool calling, structured output, reasoning. Served by 3 providers.
by amazonDec 2, 2025$0.30/M input tokens$2.50/M output tokens3 providers
- Index 13.4
Reads text. Tool calling. Served by 3 providers.
by amazonDec 1, 20251M context$0.51/M input tokens$4.25/M output tokens3 providers
- 1M context
Reads text and images. Tool calling.
by amazonOct 31, 2025$2.50/M input tokens$12.50/M output tokens
- 300K context
Reads text. Tool calling, structured output. Served by 4 providers.
by amazonDec 5, 2024$0.0595/M input tokens$0.238/M output tokens4 providers
Titan Text Embeddings V2 questions
What is the context window of Titan Text Embeddings V2?
8,192 tokens of input, and it can write up to 1,536 tokens in one reply. As a rough guide, a million tokens is around 750,000 English words.
Is Titan Text Embeddings V2 open source?
No. Amazon serves it through an API only, so you cannot download the weights or host it yourself.
When was Titan Text Embeddings V2 released?
Apr 30, 2024.
Catalog data from models.dev (MIT licensed), refreshed daily, last on Oct 15, 2026. The summary is written from these facts.
Building on Titan Text Embeddings V2?
I test your real tasks on it and the two closest alternatives, then ship the agent on whichever wins on cost and accuracy.
Book Meeting