Skip to content

feat: add DISTILUSE_BASE_MULTILINGUAL_CASED_V2 text embeddings model #1098

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
msluszniak merged 6 commits into from
Apr 28, 2026
+77 −31
Merged

feat: add DISTILUSE_BASE_MULTILINGUAL_CASED_V2 text embeddings model #1098

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
6 commits
Select commit Hold shift + click to select a range
15ddd07
feat: add DISTILUSE_BASE_MULTILINGUAL_CASED_V2 text embeddings model
msluszniak Apr 24, 2026
1a01d54
feat: expose distiluse 8da4w + CoreML fp32/fp16 variants
msluszniak Apr 24, 2026
e4a6c4d
docs: move distiluse variant sizes from useTextEmbeddings hook page t…
msluszniak Apr 24, 2026
cf144a0
docs: add DISTILUSE benchmarks for OnePlus 12 and iPhone 17 Pro
msluszniak Apr 27, 2026
e318d6a
refactor: drop fp32 suffix from CoreML default-precision variant
msluszniak Apr 28, 2026
8c71bfb
refactor: drop XNNPACK fp32 and CoreML fp16 distiluse variants
msluszniak Apr 28, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 4 additions & 0 deletions .cspell-wordlist.txt
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -53,6 +53,10 @@ Lexend
finetuned
MINILM
MPNET
DISTILUSE
distiluse
Distil
torchao
QINT
FNUZ
wordlist
Expand Down
10 changes: 10 additions & 0 deletions apps/text-embeddings/app/text-embeddings/index.tsx
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -18,6 +18,8 @@ import {
ALL_MPNET_BASE_V2,
MULTI_QA_MINILM_L6_COS_V1,
MULTI_QA_MPNET_BASE_DOT_V1,
DISTILUSE_BASE_MULTILINGUAL_CASED_V2_8DA4W,
DISTILUSE_BASE_MULTILINGUAL_CASED_V2_COREML,
TextEmbeddingsProps,
} from 'react-native-executorch';

Expand All @@ -28,6 +30,14 @@ const MODELS: { label: string; value: TextEmbeddingModel }[] = [
{ label: 'MPNet Base', value: ALL_MPNET_BASE_V2 },
{ label: 'MultiQA MiniLM', value: MULTI_QA_MINILM_L6_COS_V1 },
{ label: 'MultiQA MPNet', value: MULTI_QA_MPNET_BASE_DOT_V1 },
{
label: 'Multilingual DistilUSE (8da4w)',
value: DISTILUSE_BASE_MULTILINGUAL_CASED_V2_8DA4W,
},
{
label: 'Multilingual DistilUSE (CoreML)',
value: DISTILUSE_BASE_MULTILINGUAL_CASED_V2_COREML,
},
];
import { useIsFocused } from '@react-navigation/native';
import { dotProduct } from '../../utils/math';
Expand Down
16 changes: 9 additions & 7 deletions docs/docs/02-benchmarks/inference-time.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -150,13 +150,15 @@ Average time to synthesize speech from an input text of approximately 60 tokens,
Benchmark times for text embeddings are highly dependent on the sentence length. The numbers below are based on a sentence of around 80 tokens. For shorter or longer sentences, inference time may vary accordingly.
:::

| Model | iPhone 17 Pro (XNNPACK) [ms] | OnePlus 12 (XNNPACK) [ms] |
| -------------------------- | :--------------------------: | :-----------------------: |
| ALL_MINILM_L6_V2 | 7 | 21 |
| ALL_MPNET_BASE_V2 | 24 | 90 |
| MULTI_QA_MINILM_L6_COS_V1 | 7 | 19 |
| MULTI_QA_MPNET_BASE_DOT_V1 | 24 | 88 |
| CLIP_VIT_BASE_PATCH32_TEXT | 14 | 39 |
| Model / Device | iPhone 17 Pro [ms] | OnePlus 12 [ms] |
| ---------------------------------------------------- | :----------------: | :-------------: |
| ALL_MINILM_L6_V2 (XNNPACK) | 7 | 21 |
| ALL_MPNET_BASE_V2 (XNNPACK) | 24 | 90 |
| MULTI_QA_MINILM_L6_COS_V1 (XNNPACK) | 7 | 19 |
| MULTI_QA_MPNET_BASE_DOT_V1 (XNNPACK) | 24 | 88 |
| CLIP_VIT_BASE_PATCH32_TEXT (XNNPACK) | 14 | 39 |
| DISTILUSE_BASE_MULTILINGUAL_CASED_V2 (XNNPACK 8da4w) | 16 | 15 |
| DISTILUSE_BASE_MULTILINGUAL_CASED_V2 (Core ML FP32) | 15 | - |

## Image Embeddings

Expand Down
20 changes: 11 additions & 9 deletions docs/docs/02-benchmarks/memory-usage.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -98,13 +98,15 @@ The reported memory usage values include the memory footprint of the Phonemis pa

## Text Embeddings

| Model / Device | iPhone 17 Pro [MB] | OnePlus 12 [MB] |
| ------------------------------------ | :----------------: | :-------------: |
| ALL_MINILM_L6_V2 (XNNPACK) | 110 | 95 |
| ALL_MPNET_BASE_V2 (XNNPACK) | 455 | 405 |
| MULTI_QA_MINILM_L6_COS_V1 (XNNPACK) | 140 | 120 |
| MULTI_QA_MPNET_BASE_DOT_V1 (XNNPACK) | 455 | 435 |
| CLIP_VIT_BASE_PATCH32_TEXT (XNNPACK) | 280 | 200 |
| Model / Device | iPhone 17 Pro [MB] | OnePlus 12 [MB] |
| ---------------------------------------------------- | :----------------: | :-------------: |
| ALL_MINILM_L6_V2 (XNNPACK) | 110 | 95 |
| ALL_MPNET_BASE_V2 (XNNPACK) | 455 | 405 |
| MULTI_QA_MINILM_L6_COS_V1 (XNNPACK) | 140 | 120 |
| MULTI_QA_MPNET_BASE_DOT_V1 (XNNPACK) | 455 | 435 |
| CLIP_VIT_BASE_PATCH32_TEXT (XNNPACK) | 280 | 200 |
| DISTILUSE_BASE_MULTILINGUAL_CASED_V2 (XNNPACK 8da4w) | 36 | 44 |
| DISTILUSE_BASE_MULTILINGUAL_CASED_V2 (Core ML FP32) | 55 | - |

## Image Embeddings

Expand All @@ -120,8 +122,8 @@ output. When resize is enabled, expect higher memory usage and inference time
with higher resolutions.
:::

| Model / Device | iPhone 17 Pro [MB] | OnePlus 12 [MB] |
| --------------------------- | :----------------: | :-------------: |
| Model / Device | iPhone 17 Pro [MB] | OnePlus 12 [MB] |
| ---------------------------- | :----------------: | :-------------: |
| DEEPLABV3_RESNET50 (XNNPACK) | 660 | 930 |

## Instance Segmentation
Expand Down
16 changes: 9 additions & 7 deletions docs/docs/02-benchmarks/model-size.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -89,13 +89,15 @@ title: Model Size

## Text Embeddings

| Model | XNNPACK [MB] |
| -------------------------- | :----------: |
| ALL_MINILM_L6_V2 | 91 |
| ALL_MPNET_BASE_V2 | 438 |
| MULTI_QA_MINILM_L6_COS_V1 | 91 |
| MULTI_QA_MPNET_BASE_DOT_V1 | 438 |
| CLIP_VIT_BASE_PATCH32_TEXT | 254 |
| Model | Size [MB] |
| ------------------------------------------- | :-------: |
| ALL_MINILM_L6_V2 | 91 |
| ALL_MPNET_BASE_V2 | 438 |
| MULTI_QA_MINILM_L6_COS_V1 | 91 |
| MULTI_QA_MPNET_BASE_DOT_V1 | 438 |
| CLIP_VIT_BASE_PATCH32_TEXT | 254 |
| DISTILUSE_BASE_MULTILINGUAL_CASED_V2_8DA4W | 393 |
| DISTILUSE_BASE_MULTILINGUAL_CASED_V2_COREML | 541 |

## Image Embeddings

Expand Down
15 changes: 8 additions & 7 deletions docs/docs/03-hooks/01-natural-language-processing/useTextEmbeddings.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -101,13 +101,14 @@ function App() {

## Supported models

| Model | Language | Max Tokens | Embedding Dimensions | Description |
| ----------------------------------------------------------------------------------------------------- | :------: | :--------: | :------------------: | -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| [all-MiniLM-L6-v2](https://huggingface.co/sentence-transformers/all-MiniLM-L6-v2) | English | 254 | 384 | All-round model tuned for many use-cases. Trained on a large and diverse dataset of over 1 billion training pairs. |
| [all-mpnet-base-v2](https://huggingface.co/sentence-transformers/all-mpnet-base-v2) | English | 382 | 768 | All-round model tuned for many use-cases. Trained on a large and diverse dataset of over 1 billion training pairs. |
| [multi-qa-MiniLM-L6-cos-v1](https://huggingface.co/sentence-transformers/multi-qa-MiniLM-L6-cos-v1) | English | 509 | 384 | This model was tuned for semantic search: Given a query/question, it can find relevant passages. It was trained on a large and diverse set of (question, answer) pairs. |
| [multi-qa-mpnet-base-dot-v1](https://huggingface.co/sentence-transformers/multi-qa-mpnet-base-dot-v1) | English | 510 | 768 | This model was tuned for semantic search: Given a query/question, it can find relevant passages. It was trained on a large and diverse set of (question, answer) pairs. |
| [clip-vit-base-patch32-text](https://huggingface.co/openai/clip-vit-base-patch32) | English | 74 | 512 | CLIP (Contrastive Language-Image Pre-Training) is a neural network trained on a variety of (image, text) pairs. CLIP allows to embed images and text into the same vector space. This allows to find similar images as well as to implement image search. This is the text encoder part of the CLIP model. To embed images checkout [clip-vit-base-patch32-image](../02-computer-vision/useImageEmbeddings.md#supported-models). |
| Model | Language | Max Tokens | Embedding Dimensions | Description |
| ------------------------------------------------------------------------------------------------------------------------- | :-----------: | :--------: | :------------------: | -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| [all-MiniLM-L6-v2](https://huggingface.co/sentence-transformers/all-MiniLM-L6-v2) | English | 254 | 384 | All-round model tuned for many use-cases. Trained on a large and diverse dataset of over 1 billion training pairs. |
| [all-mpnet-base-v2](https://huggingface.co/sentence-transformers/all-mpnet-base-v2) | English | 382 | 768 | All-round model tuned for many use-cases. Trained on a large and diverse dataset of over 1 billion training pairs. |
| [multi-qa-MiniLM-L6-cos-v1](https://huggingface.co/sentence-transformers/multi-qa-MiniLM-L6-cos-v1) | English | 509 | 384 | This model was tuned for semantic search: Given a query/question, it can find relevant passages. It was trained on a large and diverse set of (question, answer) pairs. |
| [multi-qa-mpnet-base-dot-v1](https://huggingface.co/sentence-transformers/multi-qa-mpnet-base-dot-v1) | English | 510 | 768 | This model was tuned for semantic search: Given a query/question, it can find relevant passages. It was trained on a large and diverse set of (question, answer) pairs. |
| [distiluse-base-multilingual-cased-v2](https://huggingface.co/sentence-transformers/distiluse-base-multilingual-cased-v2) | 50+ languages | 126 | 512 | Multilingual DistilBERT with a 768→512 projection head. Recommended when broader language coverage matters more than the exact English quality of MiniLM/MPNet. |
| [clip-vit-base-patch32-text](https://huggingface.co/openai/clip-vit-base-patch32) | English | 74 | 512 | CLIP (Contrastive Language-Image Pre-Training) is a neural network trained on a variety of (image, text) pairs. CLIP allows to embed images and text into the same vector space. This allows to find similar images as well as to implement image search. This is the text encoder part of the CLIP model. To embed images checkout [clip-vit-base-patch32-image](../02-computer-vision/useImageEmbeddings.md#supported-models). |

**`Max Tokens`** - The maximum number of tokens that can be processed by the model. If the input text exceeds this limit, it will be truncated.

Expand Down
25 changes: 24 additions & 1 deletion packages/react-native-executorch/src/constants/modelUrls.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,5 +1,5 @@
import { Platform } from 'react-native';
import { URL_PREFIX, VERSION_TAG } from './versions';
import { URL_PREFIX, VERSION_TAG, NEXT_VERSION_TAG } from './versions';

// LLMs

Expand Down Expand Up @@ -987,6 +987,9 @@ const MULTI_QA_MINILM_L6_COS_V1_MODEL = `${URL_PREFIX}-multi-qa-MiniLM-L6-cos-v1
const MULTI_QA_MINILM_L6_COS_V1_TOKENIZER = `${URL_PREFIX}-multi-qa-MiniLM-L6-cos-v1/${VERSION_TAG}/tokenizer.json`;
const MULTI_QA_MPNET_BASE_DOT_V1_MODEL = `${URL_PREFIX}-multi-qa-mpnet-base-dot-v1/${VERSION_TAG}/multi-qa-mpnet-base-dot-v1_xnnpack.pte`;
const MULTI_QA_MPNET_BASE_DOT_V1_TOKENIZER = `${URL_PREFIX}-multi-qa-mpnet-base-dot-v1/${VERSION_TAG}/tokenizer.json`;
const DISTILUSE_BASE_MULTILINGUAL_CASED_V2_8DA4W_MODEL = `${URL_PREFIX}-distiluse-base-multilingual-cased-v2/${NEXT_VERSION_TAG}/xnnpack/distiluse-base-multilingual-cased-v2_xnnpack_8da4w.pte`;
const DISTILUSE_BASE_MULTILINGUAL_CASED_V2_COREML_MODEL = `${URL_PREFIX}-distiluse-base-multilingual-cased-v2/${NEXT_VERSION_TAG}/coreml/distiluse-base-multilingual-cased-v2_coreml_fp32.pte`;
const DISTILUSE_BASE_MULTILINGUAL_CASED_V2_TOKENIZER = `${URL_PREFIX}-distiluse-base-multilingual-cased-v2/${NEXT_VERSION_TAG}/tokenizer.json`;
const CLIP_VIT_BASE_PATCH32_TEXT_MODEL = `${URL_PREFIX}-clip-vit-base-patch32/${VERSION_TAG}/xnnpack/clip_vit_base_patch32_text_xnnpack_fp32.pte`;
const CLIP_VIT_BASE_PATCH32_TEXT_TOKENIZER = `${URL_PREFIX}-clip-vit-base-patch32/${VERSION_TAG}/tokenizer.json`;

Expand Down Expand Up @@ -1026,6 +1029,24 @@ export const MULTI_QA_MPNET_BASE_DOT_V1 = {
tokenizerSource: MULTI_QA_MPNET_BASE_DOT_V1_TOKENIZER,
} as const;

/**
* @category Models - Text Embeddings
*/
export const DISTILUSE_BASE_MULTILINGUAL_CASED_V2_8DA4W = {
modelName: 'distiluse-base-multilingual-cased-v2-8da4w',
modelSource: DISTILUSE_BASE_MULTILINGUAL_CASED_V2_8DA4W_MODEL,
tokenizerSource: DISTILUSE_BASE_MULTILINGUAL_CASED_V2_TOKENIZER,
} as const;

/**
* @category Models - Text Embeddings
*/
export const DISTILUSE_BASE_MULTILINGUAL_CASED_V2_COREML = {
modelName: 'distiluse-base-multilingual-cased-v2-coreml',
modelSource: DISTILUSE_BASE_MULTILINGUAL_CASED_V2_COREML_MODEL,
tokenizerSource: DISTILUSE_BASE_MULTILINGUAL_CASED_V2_TOKENIZER,
} as const;

/**
* @category Models - Text Embeddings
*/
Expand Down Expand Up @@ -1175,6 +1196,8 @@ export const MODEL_REGISTRY = {
ALL_MPNET_BASE_V2,
MULTI_QA_MINILM_L6_COS_V1,
MULTI_QA_MPNET_BASE_DOT_V1,
DISTILUSE_BASE_MULTILINGUAL_CASED_V2_8DA4W,
DISTILUSE_BASE_MULTILINGUAL_CASED_V2_COREML,
CLIP_VIT_BASE_PATCH32_TEXT,
BK_SDM_TINY_VPRED_512,
BK_SDM_TINY_VPRED_256,
Expand Down
2 changes: 2 additions & 0 deletions packages/react-native-executorch/src/types/textEmbeddings.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -10,6 +10,8 @@ export type TextEmbeddingsModelName =
| 'all-mpnet-base-v2'
| 'multi-qa-minilm-l6-cos-v1'
| 'multi-qa-mpnet-base-dot-v1'
| 'distiluse-base-multilingual-cased-v2-8da4w'
| 'distiluse-base-multilingual-cased-v2-coreml'
| 'clip-vit-base-patch32-text';

/**
Expand Down
Loading