Update model card

README.md

@@ -1,61 +1,68 @@
----
-language:
-- ja
----
-
-# Model Card for `answer-finder.yuzu`
-
-This model is a question answering model developed by Sinequa. It produces two lists of logit scores corresponding to
-the start token and end token of an answer.
-
-Model name: `answer-finder.yuzu`
-
-## Supported Languages
-
-The model was trained and tested in the following languages:
-
-- Japanese
-
-Besides the aforementioned languages, basic support can be expected for the 104 languages that were used during the
-pretraining of the base model (See [original repository](https://github.com/google-research/bert)).
-
-## Scores
-
-| Metric                                                        |  Value |
-|:--------------------------------------------------------------|-------:|
-| F1 Score on JSQuAD with Hugging Face evaluation pipeline      |   92.1 |
-| F1 Score on JSQuAD with Haystack evaluation pipeline          |   91.5 |
-
-## Inference Time
-
-| GPU                                                           |  Batch size 1  |  Batch size 32 |
-|:--------------------------------------------------------------|---------------:|---------------:|
-| NVIDIA A10                                                    |           4 ms |          84 ms |
-| NVIDIA T4                                                     |          15 ms |         361 ms |
-
-The inference times only measure the time the model takes to process a single batch, it does not include pre- or
-post-processing steps like the tokenization.
-
-**Note that the Answer Finder models are only used at query time.**
-
-## Requirements
-
-- Minimal Sinequa version: 11.10.0
-- GPU memory usage: 1320 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.
-
-## Model Details
-
-### Overview
-
-- Number of parameters: 110 million
-- Base language model: [bert-base-multilingual-cased](https://huggingface.co/bert-base-multilingual-cased)
-- Sensitive to casing and accents
-
-### Training Data
-
-- [JSQuAD](https://github.com/yahoojapan/JGLUE) see [Paper](https://aclanthology.org/2022.lrec-1.317.pdf)
-- Japanese translation of SQuAD v2 "impossible" query-passage pairs
+---
+language:
+- ja
+---
+
+# Model Card for `answer-finder.yuzu`
+
+This model is a question answering model developed by Sinequa. It produces two lists of logit scores corresponding to the start token and end token of an answer.
+
+Model name: `answer-finder.yuzu`
+
+## Supported Languages
+
+The model was trained and tested in the following languages:
+
+- Japanese
+
+Besides the aforementioned languages, basic support can be expected for the 104 languages that were used during the pretraining of the base model (See [original repository](https://github.com/google-research/bert)).
+
+## Scores
+
+| Metric                                                        |  Value |
+|:--------------------------------------------------------------|-------:|
+| F1 Score on JSQuAD with Hugging Face evaluation pipeline      |   92.1 |
+| F1 Score on JSQuAD with Haystack evaluation pipeline          |   91.5 |
+
+## Inference Time
+
+| GPU                                       | Quantization type |  Batch size 1  |  Batch size 32 |
+|:------------------------------------------|:------------------|---------------:|---------------:|
+| NVIDIA A10                                | FP16              |          17 ms |          27 ms |
+| NVIDIA A10                                | FP32              |           4 ms |          88 ms |
+| NVIDIA T4                                 | FP16              |           3 ms |          64 ms |
+| NVIDIA T4                                 | FP32              |          15 ms |         374 ms |
+| NVIDIA L4                                 | FP16              |           3 ms |          39 ms |
+| NVIDIA L4                                 | FP32              |           5 ms |         125 ms |
+
+**Note that the Answer Finder models are only used at query time.**
+
+## Gpu Memory usage
+
+| Quantization type                                |   Memory   |
+|:-------------------------------------------------|-----------:|
+| FP16                                             |    950 MiB |
+| FP32                                             |   1350 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: 110 million
+- Base language model: [bert-base-multilingual-cased](https://huggingface.co/bert-base-multilingual-cased)
+- Sensitive to casing and accents
+
+### Training Data
+
+- [JSQuAD](https://github.com/yahoojapan/JGLUE) see [Paper](https://aclanthology.org/2022.lrec-1.317.pdf)
+- Japanese translation of SQuAD v2 "impossible" query-passage pairs