Model Gallery - Pinecone Docs

Inference

Build end-to-end faster with models hosted by Pinecone.

Request a model

llama-text-embed-v2

NVIDIA

State of the art text embedding model

Task	Embedding
Modality	Text
Max Input Tokens	2048
Price	$0.16 / million tokens

multilingual-e5-large

MICROSOFT

Top performing text embedding model from Microsoft research.

Task	Embedding
Modality	Text
Max Input Tokens	507
Price	$0.08 / million tokens

cohere-rerank-3.5

COHERE

Cohere's latest reranker with multilingual capabilities.

Task	Rerank
Modality	Text
Max Input Tokens	4096
Price	$2.00 / 1k requests

pinecone-sparse-english-v0

PINECONE

Sparse vector model for keyword-style search.

Task	Embedding
Modality	Text
Max Input Tokens	512
Price	$0.08 / million tokens

bge-reranker-v2-m3

BAAI

High performance multi-lingual reranker model.

Task	Rerank
Modality	Text
Max Input Tokens	1024
Price	$2.00 / 1k requests