Cohere’s Embed Models (Details and Application)
Cohere’s Embed Models (Details and Application)
Embed models can be used to generate embeddings from text or classify it based on various parameters. Embeddings can be used for estimating semantic similarity between two texts, choosing a sentence which is most likely to follow another sentence, or categorizing user feedback. When used with the Classify endpoint, embeddings can be used for any classification or analysis task.
Embed 5
Embed 5 is our latest embedding model family, available in two tiers. Pro and Fast share an embedding space, so a corpus indexed with one model can be queried with the other. We recommend indexing with Pro and querying with Fast.
embed-v5.0-pro: Highest retrieval quality. Best for offline indexing and quality-critical retrieval.embed-v5.0-fast: Best for low-latency, high-throughput workloads such as interactive search, agent loops, and high-volume query traffic.
Model details
- Context length: 128k tokens
- Number of parameters: ~2.4B (Pro), ~1B (Fast)
- Input: Text, images, and mixed text-and-image inputs (e.g. PDF pages)
- Output: Matryoshka embeddings with dimensions of
[256, 512, 768, 1024, 1536, 2048 (default)], asfloat,int8, orbinary - Languages: 100+
- Endpoint: Embed
List of Supported Languages
Our multilingual embed model supports over 100 languages, including Chinese, Spanish, and French.
Frequently Asked Questions
What is the Context Length for Cohere Embeddings Models?
You can find the context length for various Cohere embeddings models in the tables above. It’s in the “Max Tokens (Context Length)” column.