| Model | Architecture | Params | Task | Quality | Throughput | Latency |
|---|---|---|---|---|---|---|
| vidore/colqwen2.5-v0.2 Multi-Vec | Qwen2 | 7.0B | Encode | 0.7680 | 2 tps | 1.9s |
| vidore/colpali-v1.3-hf Multi-Vec | PaliGemma | 3.0B | Encode | 0.7119 | 6 tps | 619.1ms |
| Alibaba-NLP/gte-Qwen2-1.5B-instruct Dense | Qwen2 | 1.8B | Encode | 0.6524 | 12.3K tps | 261.1ms |
| NovaSearch/stella_en_1.5B_v5 Dense | Qwen2 | 1.5B | Encode | 0.9134 | 12.8K tps | 265.9ms |
| laion/CLIP-ViT-H-14-laion2B-s32B-b79K Dense | CLIP | 986M | Encode | 0.8624 | 321 tps | 503.8ms |
| google/siglip-so400m-patch14-384 Dense | SigLIP | 878M | Encode | 0.9001 | 355 tps | 488.3ms |
| google/siglip-so400m-patch14-224 Dense | SigLIP | 877M | Encode | 0.8382 | 348 tps | 439.8ms |
| Qwen/Qwen3-Embedding-0.6B Dense | Qwen3 | 596M | Encode | 0.6538 | 20.6K tps | 156.9ms |
| BAAI/bge-m3 DenseSparseMulti-Vec | XLM-RoBERTa | 568M | Encode | 0.5726 | 32.3K tps | 94.1ms |
| BAAI/bge-reranker-large Score | XLM-RoBERTa | 560M | Score | 0.6404 | 6.6K tps | 41.4ms |
| intfloat/multilingual-e5-large Dense | XLM-RoBERTa | 560M | Encode | 0.5035 | 29.8K tps | 108.6ms |
| intfloat/multilingual-e5-large-instruct Dense | XLM-RoBERTa | 560M | Encode | 0.5539 | 29.4K tps | 106.9ms |
| jinaai/jina-colbert-v2 Multi-Vec | XLM-RoBERTa | 559M | Encode | 0.7615 | 24.9K tps | 146.1ms |
| nomic-ai/nomic-embed-text-v2-moe Dense | NomicBERT | 475M | Encode | 0.5207 | 13.0K tps | 149.6ms |
| numind/NuNER_Zero Entities | DeBERTa | 449M | Extract | 0.6122 | — | — |
| NovaSearch/stella_en_400M_v5 Dense | ModernBERT | 435M | Encode | 0.8666 | 27.1K tps | 115.7ms |
| EmergentMethods/gliner_large_news-v2.1 Entities | DeBERTa | 435M | Extract | 0.5527 | — | — |
| Ihor/gliner-biomed-large-v1.0 Entities | DeBERTa | 435M | Extract | 0.6439 | — | — |
| jackboyla/glirel-large-v0 Relations | DeBERTa | 435M | Extract | 0.2639 | — | — |
| mixedbread-ai/mxbai-rerank-large-v2 Score | Qwen2 | 435M | Score | 0.6914 | 2.3K tps | 1.1s |
| urchade/gliner_large-v2.1 Entities | DeBERTa | 435M | Extract | 0.5483 | — | — |
| urchade/gliner_multi-v2.1 Entities | DeBERTa | 435M | Extract | 0.6007 | — | — |
| urchade/gliner_multi_pii-v1 Entities | DeBERTa | 435M | Extract | 0.5357 | — | — |
| openai/clip-vit-large-patch14 Dense | CLIP | 428M | Encode | 0.7824 | 706 tps | 298.1ms |
| mixedbread-ai/mxbai-colbert-large-v1 Multi-Vec | BERT | 335M | Encode | 0.4833 | 42.7K tps | 77.3ms |
| intfloat/e5-large-v2 Dense | BERT | 335M | Encode | 0.4531 | 33.2K tps | 86.6ms |
| Alibaba-NLP/gte-multilingual-base Dense | ModernBERT | 305M | Encode | 0.5990 | 55.1K tps | 63.1ms |
| lightonai/GTE-ModernColBERT-v1 Multi-Vec | ModernBERT | 305M | Encode | 0.7773 | 5.3K tps | 355.9ms |
| google/embeddinggemma-300m Dense | Gemma 3 | 303M | Encode | 0.3876 | 79.6K tps | 55.7ms |
| jinaai/jina-reranker-v2-base-multilingual Score | XLM-RoBERTa | 278M | Score | 0.6546 | 8.3K tps | 32.0ms |
| BAAI/bge-reranker-base Score | XLM-RoBERTa | 278M | Score | 0.5926 | 5.0K tps | 33.2ms |
| IDEA-Research/grounding-dino-base Bounding Boxes | Swin | 233M | Extract | 0.5809 | 0.8 mpix/s | 785.8ms |
| urchade/gliner_medium-v2.1 Entities | DeBERTa | 195M | Extract | 0.6111 | — | — |
| IDEA-Research/grounding-dino-tiny Bounding Boxes | Swin | 172M | Extract | 0.4860 | 0.9 mpix/s | 532.6ms |
| google/owlv2-base-patch16-ensemble Bounding Boxes | CLIP | 155M | Extract | 0.5171 | 1.0 mpix/s | 954.6ms |
| laion/CLIP-ViT-B-32-laion2B-s34B-b79K Dense | CLIP | 151M | Encode | 0.7744 | 1.2K tps | 178.6ms |
| openai/clip-vit-base-patch32 Dense | CLIP | 151M | Encode | 0.7165 | 651 tps | 319.4ms |
| mixedbread-ai/mxbai-rerank-base-v2 Score | Qwen2 | 150M | Score | 0.6638 | 6.0K tps | 454.0ms |
| Alibaba-NLP/gte-reranker-modernbert-base Score | ModernBERT | 150M | Score | 0.6701 | 6.2K tps | 41.9ms |
| lightonai/Reason-ModernColBERT Multi-Vec | ModernBERT | 149M | Encode | 0.7777 | 33.0K tps | 82.2ms |
| opensearch-project/opensearch-neural-sparse-encoding-doc-v3-gte Sparse | ModernBERT | 137M | Encode | 0.7470 | 34.2K tps | 93.7ms |
| naver/splade-cocondenser-selfdistil Sparse | BERT | 110M | Encode | 0.3403 | 40.0K tps | 72.4ms |
| naver/splade-v3 Sparse | BERT | 110M | Encode | 0.7393 | 29.6K tps | 83.7ms |
| numind/NuNER_Zero-span Entities | DeBERTa | 110M | Extract | 0.6448 | — | — |
| intfloat/e5-base-v2 Dense | BERT | 109M | Encode | 0.4603 | 53.2K tps | 57.9ms |
| opensearch-project/opensearch-neural-sparse-encoding-doc-v2-distill Sparse | DistilBERT | 67M | Encode | 0.3398 | 49.1K tps | 63.3ms |
| opensearch-project/opensearch-neural-sparse-encoding-doc-v3-distill Sparse | DistilBERT | 67M | Encode | 0.3399 | 50.1K tps | 60.7ms |
| opensearch-project/opensearch-neural-sparse-encoding-v2-distill Sparse | DistilBERT | 67M | Encode | 0.3373 | 44.2K tps | 63.3ms |
| urchade/gliner_small-v2.1 Entities | DeBERTa | 60M | Extract | 0.5959 | — | — |
| answerdotai/answerai-colbert-small-v1 Multi-Vec | BERT | 33M | Encode | 0.7840 | 43.6K tps | 60.2ms |
| cross-encoder/ms-marco-MiniLM-L-12-v2 Score | BERT | 33M | Score | 0.6145 | 8.2K tps | 31.7ms |
| intfloat/e5-small-v2 Dense | BERT | 33M | Encode | 0.4299 | 58.3K tps | 49.7ms |
| mixedbread-ai/mxbai-edge-colbert-v0-32m Multi-Vec | ModernBERT | 32M | Encode | 0.3376 | 45.9K tps | 59.7ms |
| sentence-transformers/all-MiniLM-L6-v2 Dense | BERT | 23M | Encode | 0.8396 | 55.3K tps | 53.3ms |
| rasyosef/splade-mini Sparse | BERT | 11M | Encode | 0.3090 | 56.3K tps | 56.0ms |
