Reranker Models

Supported Models

Model	Max Length	Notes
BAAI/bge-reranker-v2-m3	8192	Multilingual
jinaai/jina-reranker-v2-base-multilingual	1024	Multilingual
Alibaba-NLP/gte-reranker-modernbert-base	8192	ModernBERT architecture
cross-encoder/ms-marco-MiniLM-L-12-v2	512	Smaller, faster

See Full model catalog for the complete list.

Model Selection

By Language Support

English only:

• BAAI/bge-reranker-base, BAAI/bge-reranker-large
• cross-encoder/ms-marco-MiniLM-L-6-v2, cross-encoder/ms-marco-MiniLM-L-12-v2

Multilingual (100+ languages):

• BAAI/bge-reranker-v2-m3
• jinaai/jina-reranker-v2-base-multilingual
• jinaai/jina-colbert-v2

By Context Length

Short context (512 tokens):

• BAAI/bge-reranker-base, BAAI/bge-reranker-large
• cross-encoder/ms-marco-MiniLM-L-*
• colbert-ir/colbertv2.0, mixedbread-ai/mxbai-colbert-large-v1

Medium context (1024 tokens):

• jinaai/jina-reranker-v2-base-multilingual

Long context (8192 tokens):

• BAAI/bge-reranker-v2-m3
• Alibaba-NLP/gte-reranker-modernbert-base
• mixedbread-ai/mxbai-rerank-base-v2, mixedbread-ai/mxbai-rerank-large-v2
• jinaai/jina-colbert-v2, lightonai/GTE-ModernColBERT-v1, lightonai/Reason-ModernColBERT

By Size

Compact (fast inference):

• cross-encoder/ms-marco-MiniLM-L-6-v2 - smallest cross-encoder
• mixedbread-ai/mxbai-edge-colbert-v0-32m - 32M parameters
• answerdotai/answerai-colbert-small-v1 - compact ColBERT

Large (higher capacity):

• BAAI/bge-reranker-large
• mixedbread-ai/mxbai-rerank-large-v2
• mixedbread-ai/mxbai-colbert-large-v1
• lightonai/Reason-ModernColBERT - long context (8192), ModernBERT family

Benchmarking

The best reranker for your workload is the one that wins on your data. See Evals for how to measure quality and performance, and the retrieval-ablation example in the SIE repository for a complete benchmark comparing rerankers and retrieval strategies end to end:

cd examples/retrieval-ablation
uv sync

# Validate config (no GPU needed)
uv run python benchmark_ablation.py --dry-run

What’s Next

• Multi-vector reranking - ColBERT MaxSim scoring
• Full model catalog - all supported models