138 Multilingual Tasks Evaluated: Thoroughly evaluated across ViDoRe V1, V2, V3, and JinaVDR across 4 metric families (nDCG, Recall, MAP, MRR @1/5/10). S(Q, D) = \sum{i=1}^{|Q|} \max{j=1}^{|D|} (qi \cdot dj) - HAC Token Compression (Hierarchical Agglomerative Clustering): A plug-and-play, training-free token reduction algorithm that aggregates visual patch tokens into 32 or 64 semantic centroids in joint feature-spatial space, reducing 1M-page index footprints to as little as 3.81 GiB. Official ranking on the ViDoRe leaderboard (ViDoRe V3, Mean Task). Performance comparison across modern multi-vector late-interaction visual document retrievers on ViDoRe: EVIE-4.5B embeds document and query…
SAVRN Model Hub · Models by Task
Visual Document Retrieval Models
1 open-weight visual document retrieval models in the SAVRN Model Hub, with Tencent publishing the most.
1Models
1Publishers
4.5B to 4.5BParameter range
1Licenses
Most Downloaded
| Model | Publisher | Parameters | License | Monthly downloads | Cheapest GPUs at 16-bit |
|---|---|---|---|---|---|
| EVIE-4.5B | Tencent | 4.5B | apache-2.0 | 2k | 1x MI300X, $1.85/hr |
Licenses
| License | Models | Commercial use |
|---|---|---|
| apache-2.0 | 1 | Yes |
Who Publishes Them
| Publisher | Models |
|---|---|
| Tencent | 1 |
All 1 Models
Questions
Which Visual document retrieval models are most downloaded?
By monthly downloads reported by the Hugging Face Hub: EVIE-4.5B (2k).