Skip to content

chore: Update non-LLM benchmarks post C++ port #678

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

Open
wants to merge 4 commits into
base: main
Choose a base branch
from
Open

chore: Update non-LLM benchmarks post C++ port #678

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
b3e10bc
chore: update v0.5.0 benchmarks (#481)
IgorSwat Oct 29, 2025
8fbfd00
chore: update v0.4.0 benchmarks (#481)
IgorSwat Nov 3, 2025
b9de78e
chore: minor naming fixes
IgorSwat Nov 3, 2025
bbd7f53
chore: update docs subsections
IgorSwat Nov 17, 2025
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
30 changes: 15 additions & 15 deletions docs/docs/02-hooks/01-natural-language-processing/useSpeechToText.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -75,20 +75,20 @@ For more information on loading resources, take a look at [loading models](../..

### Returns

| Field | Type | Description |
| --------------------------- | ---------------------------------------------------------------------------------------------------- | -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| `transcribe` | `(waveform: Float32Array \| number[], options?: DecodingOptions \| undefined) => Promise<string>` | Starts a transcription process for a given input array, which should be a waveform at 16kHz. The second argument is an options object, e.g. `{ language: 'es' }` for multilingual models. Resolves a promise with the output transcription when the model is finished. Passing `number[]` is deprecated. |
| `stream` | `(options?: DecodingOptions \| undefined) => Promise<string>` | Starts a streaming transcription process. Use in combination with `streamInsert` to feed audio chunks and `streamStop` to end the stream. The argument is an options object, e.g. `{ language: 'es' }` for multilingual models. Updates `committedTranscription` and `nonCommittedTranscription` as transcription progresses. |
| `streamInsert` | `(waveform: Float32Array \| number[]) => void` | Inserts a chunk of audio data (sampled at 16kHz) into the ongoing streaming transcription. Call this repeatedly as new audio data becomes available. Passing `number[]` is deprecated. |
| `streamStop` | `() => void` | Stops the ongoing streaming transcription process. |
| `encode` | `(waveform: Float32Array \| number[]) => Promise<Float32Array>` | Runs the encoding part of the model on the provided waveform. Passing `number[]` is deprecated. |
| `decode` | `(tokens: number[] \| Int32Array, encoderOutput: Float32Array \| number[]) => Promise<Float32Array>` | Runs the decoder of the model. Passing `number[]` is deprecated. |
| `committedTranscription` | `string` | Contains the part of the transcription that is finalized and will not change. Useful for displaying stable results during streaming. |
| `nonCommittedTranscription` | `string` | Contains the part of the transcription that is still being processed and may change. Useful for displaying live, partial results during streaming. |
| `error` | `string \| null` | Contains the error message if the model failed to load. |
| `isGenerating` | `boolean` | Indicates whether the model is currently processing an inference. |
| `isReady` | `boolean` | Indicates whether the model has successfully loaded and is ready for inference. |
| `downloadProgress` | `number` | Tracks the progress of the model download process. |
| Field | Type | Description |
| --------------------------- | ---------------------------------------------------------------------------------------------------- | ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| `transcribe` | `(waveform: Float32Array \| number[], options?: DecodingOptions \| undefined) => Promise<string>` | Starts a transcription process for a given input array, which should be a waveform at 16kHz. The second argument is an options object, e.g. `{ language: 'es' }` for multilingual models. Resolves a promise with the output transcription when the model is finished. Passing `number[]` is deprecated. |
| `stream` | `(options?: DecodingOptions \| undefined) => Promise<string>` | Starts a streaming transcription process. Use in combination with `streamInsert` to feed audio chunks and `streamStop` to end the stream. The argument is an options object, e.g. `{ language: 'es' }` for multilingual models. Updates `committedTranscription` and `nonCommittedTranscription` as transcription progresses. |
| `streamInsert` | `(waveform: Float32Array \| number[]) => void` | Inserts a chunk of audio data (sampled at 16kHz) into the ongoing streaming transcription. Call this repeatedly as new audio data becomes available. Passing `number[]` is deprecated. |
| `streamStop` | `() => void` | Stops the ongoing streaming transcription process. |
| `encode` | `(waveform: Float32Array \| number[]) => Promise<Float32Array>` | Runs the encoding part of the model on the provided waveform. Passing `number[]` is deprecated. |
| `decode` | `(tokens: number[] \| Int32Array, encoderOutput: Float32Array \| number[]) => Promise<Float32Array>` | Runs the decoder of the model. Passing `number[]` is deprecated. |
| `committedTranscription` | `string` | Contains the part of the transcription that is finalized and will not change. Useful for displaying stable results during streaming. |
| `nonCommittedTranscription` | `string` | Contains the part of the transcription that is still being processed and may change. Useful for displaying live, partial results during streaming. |
| `error` | `string \| null` | Contains the error message if the model failed to load. |
| `isGenerating` | `boolean` | Indicates whether the model is currently processing an inference. |
| `isReady` | `boolean` | Indicates whether the model has successfully loaded and is ready for inference. |
| `downloadProgress` | `number` | Tracks the progress of the model download process. |

<details>
<summary>Type definitions</summary>
Expand Down Expand Up @@ -340,4 +340,4 @@ function App() {

| Model | Android (XNNPACK) [MB] | iOS (XNNPACK) [MB] |
| ------------ | :--------------------: | :----------------: |
| WHISPER_TINY | 900 | 600 |
| WHISPER_TINY | 410 | 375 |
24 changes: 12 additions & 12 deletions docs/docs/02-hooks/01-natural-language-processing/useTextEmbeddings.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -133,25 +133,25 @@ For the supported models, the returned embedding vector is normalized, meaning t

| Model | Android (XNNPACK) [MB] | iOS (XNNPACK) [MB] |
| -------------------------- | :--------------------: | :----------------: |
| ALL_MINILM_L6_V2 | 85 | 100 |
| ALL_MPNET_BASE_V2 | 390 | 465 |
| MULTI_QA_MINILM_L6_COS_V1 | 115 | 130 |
| MULTI_QA_MPNET_BASE_DOT_V1 | 415 | 490 |
| CLIP_VIT_BASE_PATCH32_TEXT | 195 | 250 |
| ALL_MINILM_L6_V2 | 95 | 110 |
| ALL_MPNET_BASE_V2 | 405 | 455 |
| MULTI_QA_MINILM_L6_COS_V1 | 120 | 140 |
| MULTI_QA_MPNET_BASE_DOT_V1 | 435 | 455 |
| CLIP_VIT_BASE_PATCH32_TEXT | 200 | 280 |

### Inference time

:::warning warning
Times presented in the tables are measured as consecutive runs of the model. Initial run times may be up to 2x longer due to model loading and initialization.
:::

| Model | iPhone 16 Pro (XNNPACK) [ms] | iPhone 14 Pro Max (XNNPACK) [ms] | iPhone SE 3 (XNNPACK) [ms] | Samsung Galaxy S24 (XNNPACK) | OnePlus 12 (XNNPACK) [ms] |
| -------------------------- | :--------------------------: | :------------------------------: | :------------------------: | :--------------------------: | :-----------------------: |
| ALL_MINILM_L6_V2 | 15 | 22 | 23 | 36 | 31 |
| ALL_MPNET_BASE_V2 | 71 | 96 | 101 | 112 | 105 |
| MULTI_QA_MINILM_L6_COS_V1 | 15 | 22 | 23 | 36 | 31 |
| MULTI_QA_MPNET_BASE_DOT_V1 | 71 | 95 | 100 | 112 | 105 |
| CLIP_VIT_BASE_PATCH32_TEXT | 31 | 47 | 48 | 55 | 49 |
| Model | iPhone 17 Pro (XNNPACK) [ms] | iPhone 16 Pro (XNNPACK) [ms] | iPhone SE 3 (XNNPACK) [ms] | Samsung Galaxy S24 (XNNPACK) [ms] | OnePlus 12 (XNNPACK) [ms] |
| -------------------------- | :--------------------------: | :--------------------------: | :------------------------: | :-------------------------------: | :-----------------------: |
| ALL_MINILM_L6_V2 | 16 | 16 | 19 | 54 | 28 |
| ALL_MPNET_BASE_V2 | 115 | 116 | 144 | 145 | 95 |
| MULTI_QA_MINILM_L6_COS_V1 | 16 | 16 | 20 | 47 | 28 |
| MULTI_QA_MPNET_BASE_DOT_V1 | 112 | 119 | 144 | 146 | 96 |
| CLIP_VIT_BASE_PATCH32_TEXT | 47 | 45 | 57 | 65 | 48 |

:::info
Benchmark times for text embeddings are highly dependent on the sentence length. The numbers above are based on a sentence of around 80 tokens. For shorter or longer sentences, inference time may vary accordingly.
Expand Down
6 changes: 3 additions & 3 deletions docs/docs/02-hooks/02-computer-vision/useClassification.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -100,14 +100,14 @@ function App() {

| Model | Android (XNNPACK) [MB] | iOS (Core ML) [MB] |
| ----------------- | :--------------------: | :----------------: |
| EFFICIENTNET_V2_S | 130 | 85 |
| EFFICIENTNET_V2_S | 230 | 87 |

### Inference time

:::warning warning
Times presented in the tables are measured as consecutive runs of the model. Initial run times may be up to 2x longer due to model loading and initialization.
:::

| Model | iPhone 16 Pro (Core ML) [ms] | iPhone 13 Pro (Core ML) [ms] | iPhone SE 3 (Core ML) [ms] | Samsung Galaxy S24 (XNNPACK) [ms] | OnePlus 12 (XNNPACK) [ms] |
| Model | iPhone 17 Pro (Core ML) [ms] | iPhone 16 Pro (Core ML) [ms] | iPhone SE 3 (Core ML) [ms] | Samsung Galaxy S24 (XNNPACK) [ms] | OnePlus 12 (XNNPACK) [ms] |
| ----------------- | :--------------------------: | :--------------------------: | :------------------------: | :-------------------------------: | :-----------------------: |
| EFFICIENTNET_V2_S | 100 | 120 | 130 | 180 | 170 |
| EFFICIENTNET_V2_S | 105 | 110 | 149 | 299 | 227 |
6 changes: 3 additions & 3 deletions docs/docs/02-hooks/02-computer-vision/useImageEmbeddings.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -123,9 +123,9 @@ For the supported models, the returned embedding vector is normalized, meaning t
Times presented in the tables are measured as consecutive runs of the model. Initial run times may be up to 2x longer due to model loading and initialization. Performance also heavily depends on image size, because resize is expansive operation, especially on low-end devices.
:::

| Model | iPhone 16 Pro (XNNPACK) [ms] | iPhone 14 Pro Max (XNNPACK) [ms] | iPhone SE 3 (XNNPACK) [ms] | Samsung Galaxy S24 (XNNPACK) [ms] | OnePlus 12 (XNNPACK) [ms] |
| --------------------------- | :--------------------------: | :------------------------------: | :------------------------: | :-------------------------------: | :-----------------------: |
| CLIP_VIT_BASE_PATCH32_IMAGE | 48 | 64 | 69 | 65 | 63 |
| Model | iPhone 17 Pro (XNNPACK) [ms] | iPhone 16 Pro (XNNPACK) [ms] | iPhone SE 3 (XNNPACK) [ms] | Samsung Galaxy S24 (XNNPACK) [ms] | OnePlus 12 (XNNPACK) [ms] |
| --------------------------- | :--------------------------: | :--------------------------: | :------------------------: | :-------------------------------: | :-----------------------: |
| CLIP_VIT_BASE_PATCH32_IMAGE | 70 | 70 | 90 | 66 | 58 |

:::info
Image embedding benchmark times are measured using 224×224 pixel images, as required by the model. All input images, whether larger or smaller, are resized to 224×224 before processing. Resizing is typically fast for small images but may be noticeably slower for very large images, which can increase total inference time.
Expand Down
Loading