---
language:
  - de
  - en
  - es
  - fr
  - it
  - ja
  - nl
  - pt
  - zh
---

# Model Card for `passage-ranker.mango`

This model is a passage ranker developed by Sinequa. It produces a relevance score given a query-passage pair and is used to order search results.

Model name: `passage-ranker.mango`

## Supported Languages

The model was trained and tested in the following languages:

- Chinese (simplified)
- Dutch
- English
- French
- German
- Italian
- Japanese
- Portuguese
- Spanish

Besides the aforementioned languages, basic support can be expected for additional 93 languages that were used during the pretraining of the base model (see
[list of languages](https://github.com/google-research/bert/blob/master/multilingual.md#list-of-languages)).

## Scores

| Metric              | Value |
|:--------------------|------:|
| Relevance (NDCG@10) | 0.480 |

Note that the relevance score is computed as an average over 14 retrieval datasets (see
[details below](#evaluation-metrics)).

## Inference Times

| GPU                                       | Quantization type |  Batch size 1  |  Batch size 32 |
|:------------------------------------------|:------------------|---------------:|---------------:|
| NVIDIA A10                                | FP16              |           2 ms |          28 ms |
| NVIDIA A10                                | FP32              |           4 ms |          82 ms |
| NVIDIA T4                                 | FP16              |           3 ms |          65 ms |
| NVIDIA T4                                 | FP32              |          14 ms |         369 ms |
| NVIDIA L4                                 | FP16              |           3 ms |          38 ms |
| NVIDIA L4                                 | FP32              |           5 ms |         123 ms |

## Gpu Memory usage

| Quantization type                                |   Memory   |
|:-------------------------------------------------|-----------:|
| FP16                                             |    850 MiB |
| FP32                                             |   1200 MiB |

Note that GPU memory usage only includes how much GPU memory the actual model consumes on an NVIDIA T4 GPU with a batch
size of 32. It does not include the fix amount of memory that is consumed by the ONNX Runtime upon initialization which
can be around 0.5 to 1 GiB depending on the used GPU.

## Requirements

- Minimal Sinequa version: 11.10.0
- Minimal Sinequa version for using FP16 models and GPUs with CUDA compute capability of 8.9+ (like NVIDIA L4): 11.11.0
- [Cuda compute capability](https://developer.nvidia.com/cuda-gpus): above 5.0 (above 6.0 for FP16 use)

## Model Details

### Overview

- Number of parameters: 167 million
- Base language model: [Multilingual BERT-Base](https://huggingface.co/bert-base-multilingual-uncased)
- Insensitive to casing and accents
- Training procedure: [MonoBERT](https://arxiv.org/abs/1901.04085)

### Training Data

- MS MARCO Passage Ranking
  ([Paper](https://arxiv.org/abs/1611.09268),
  [Official Page](https://microsoft.github.io/msmarco/),
  [English & translated datasets on the HF dataset hub](https://huggingface.co/datasets/unicamp-dl/mmarco))
    - Original English dataset
    - Translated datasets for the other eight supported languages

### Evaluation Metrics

To determine the relevance score, we averaged the results that we obtained when evaluating on the datasets of the
[BEIR benchmark](https://github.com/beir-cellar/beir). Note that all these datasets are in English.

| Dataset           | NDCG@10 |
|:------------------|--------:|
| Average           |   0.480 |
|                   |         |
| Arguana           |   0.537 |
| CLIMATE-FEVER     |   0.241 |
| DBPedia Entity    |   0.371 |
| FEVER             |   0.777 |
| FiQA-2018         |   0.327 |
| HotpotQA          |   0.696 |
| MS MARCO          |   0.414 |
| NFCorpus          |   0.332 |
| NQ                |   0.484 |
| Quora             |   0.768 |
| SCIDOCS           |   0.143 |
| SciFact           |   0.648 |
| TREC-COVID        |   0.673 |
| Webis-Touche-2020 |   0.310 |

We evaluated the model on the datasets of the [MIRACL benchmark](https://github.com/project-miracl/miracl) to test its multilingual capacities. Note that not all training languages are part of the benchmark, so we only report the metrics for the existing languages.

| Language              | NDCG@10 |
|:----------------------|--------:|
| Chinese (simplified)  |   0.463 |
| French                |   0.447 |
| German                |   0.415 |
| Japanese              |   0.526 |
| Spanish               |   0.485 |