Embedding and Reranking Models

Explore currently public embedding and reranking models for vector retrieval, semantic search, and RAG on Infistar.

Currently public models

12 models
  • text-embedding-v4From Input ¥0.5 / Output ¥0 per 1M tokens
  • qwen3.7-text-embeddingFrom Input ¥0.5 / Output ¥0.5 per 1M tokens, cache read ¥0.5
  • qwen3-vl-rerankFrom Input ¥0.5 / Output ¥0.5 per 1M tokens, cache read ¥0.5
  • qwen3-rerankFrom Input ¥0.5 / Output ¥0.5 per 1M tokens, cache read ¥0.5
  • embedding-3From Input ¥0.5 / Output ¥0.5 per 1M tokens, cache read ¥0.5
  • embedding-2From Input ¥0.5 / Output ¥0.5 per 1M tokens, cache read ¥0.5
  • rerankFrom Input ¥0.8 / Output ¥0.8 per 1M tokens, cache read ¥0.8
  • baai-bge-m3From Input ¥0.01 / Output ¥0.01 per 1M tokens, cache read ¥0.01
  • baai-bge-large-zh-v1.5From Input ¥0.01 / Output ¥0.01 per 1M tokens, cache read ¥0.01
  • baai-bge-large-en-v1.5From Input ¥0.01 / Output ¥0.01 per 1M tokens, cache read ¥0.01
  • baai-bge-reranker-v2-m3From Input ¥0.02 / Output ¥0.02 per 1M tokens, cache read ¥0.02
  • gte-rerank-v2From Input ¥0.8 / Output ¥0 per 1M tokens