https://github.com/huggingface/swift-transformers
Swift Package to implement a transformers-like API in Swift
https://github.com/huggingface/swift-transformers
Last synced: 2 months ago
JSON representation
Swift Package to implement a transformers-like API in Swift
- Host: GitHub
- URL: https://github.com/huggingface/swift-transformers
- Owner: huggingface
- License: apache-2.0
- Created: 2023-05-11T10:56:48.000Z (about 3 years ago)
- Default Branch: main
- Last Pushed: 2025-09-23T16:46:14.000Z (8 months ago)
- Last Synced: 2025-09-23T18:09:27.608Z (8 months ago)
- Language: Swift
- Size: 5.04 MB
- Stars: 1,103
- Watchers: 30
- Forks: 137
- Open Issues: 21
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
- awesome-mobile-llm - HF Swift Transformers - like API in Swift (Infrastructure / Deployment of LLMs on Device / Deployment Frameworks)
README
[](https://github.com/huggingface/swift-transformers/actions/workflows/unit-tests.yml)
[](https://swiftpackageindex.com/huggingface/swift-transformers)
[](https://swiftpackageindex.com/huggingface/swift-transformers)
`swift-transformers` is a collection of utilities to help adopt language models in Swift apps.
Those familiar with the [`transformers`](https://github.com/huggingface/transformers) Python library will find a familiar yet idiomatic Swift API.
## Rationale & Overview
Check out [our v1.0 release post](https://huggingface.co/blog/swift-transformers) and our [original announcement](https://huggingface.co/blog/swift-coreml-llm) for more context on why we built this library.
## Examples
The most commonly used modules from `swift-transformers` are `Tokenizers` and `Hub`, which allow fast tokenization and
model downloads from the Hugging Face Hub.
### Tokenizing text + chat templating
Tokenizing text should feel very familiar to those who have used the Python `transformers` library:
```swift
let tokenizer = try await AutoTokenizer.from(pretrained: "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B")
let messages = [["role": "user", "content": "Describe the Swift programming language."]]
let encoded = try tokenizer.applyChatTemplate(messages: messages)
let decoded = tokenizer.decode(tokens: encoded)
```
### Tool calling
`swift-transformers` natively supports formatting inputs for tool calling, allowing for complex interactions with language models:
```swift
let tokenizer = try await AutoTokenizer.from(pretrained: "mlx-community/Qwen2.5-7B-Instruct-4bit")
let weatherTool = [
"type": "function",
"function": [
"name": "get_current_weather",
"description": "Get the current weather in a given location",
"parameters": [
"type": "object",
"properties": ["location": ["type": "string", "description": "City and state"]],
"required": ["location"]
]
]
]
let tokens = try tokenizer.applyChatTemplate(
messages: [["role": "user", "content": "What's the weather in Paris?"]],
tools: [weatherTool]
)
```
### Hub downloads
Downloading models to a user device _fast_ and _reliably_ is a core requirement of on-device ML. `swift-transformers` provides a simple API to
download models from the Hugging Face Hub, with progress reporting, flaky connection handling, and more:
```swift
let repo = Hub.Repo(id: "mlx-community/Qwen2.5-0.5B-Instruct-2bit-mlx")
let modelDirectory: URL = try await Hub.snapshot(
from: repo,
matching: ["config.json", "*.safetensors"],
progressHandler: { progress in
print("Download progress: \(progress.fractionCompleted * 100)%")
}
)
print("Files downloaded to: \(modelDirectory.path)")
```
### CoreML Integration
The `Models` and `Generation` modules provide handy utilities when working with language models in CoreML. Check out our
example converting and running Mistral 7B using CoreML [here](https://github.com/huggingface/swift-transformers/tree/main/Examples).
The [modernization of Core ML](https://github.com/huggingface/swift-transformers/pull/257) and corresponding examples were primarily contributed by @joshnewnham, @1duo, @alejandro-isaza, @aseemw. Thank you 🙏
### Offline CoreML tokenizers
When you bundle a compiled CoreML model and tokenizer files with your app, you can skip any network requests by injecting
the tokenizer when constructing `LanguageModel`:
```swift
let compiledURL: URL = ... // path to .mlmodelc
let tokenizerFolder: URL = ... // folder containing tokenizer_config.json and tokenizer.json
// Construct the tokenizer from local files (inside an async context)
let tokenizer = try await AutoTokenizer.from(modelFolder: tokenizerFolder)
let model = try LanguageModel.loadCompiled(
url: compiledURL,
tokenizer: tokenizer
)
```
Make sure the tokenizer assets come from the same Hugging Face repo as the original checkpoint or are compatible with the model you use. For the
Mistral example in `Examples/Mistral7B/`, you can fetch the tokenizer like this:
```bash
huggingface-cli download \
mistralai/Mistral-7B-Instruct-v0.3 \
tokenizer.json tokenizer_config.json \
--local-dir Examples/Mistral7B/local-tokenizer
```
If the repo is gated, authenticate with `huggingface-cli login` first. Both initializers reuse the tokenizer
you pass in and never reach out to the Hugging Face Hub.
## Usage via SwiftPM
To use `swift-transformers` with SwiftPM, you can add this to your `Package.swift`:
```swift
dependencies: [
.package(url: "https://github.com/huggingface/swift-transformers", from: "1.3.0")
]
```
And then, add the Transformers library as a dependency to your target:
```swift
targets: [
.target(
name: "YourTargetName",
dependencies: [
.product(name: "Transformers", package: "swift-transformers")
]
)
]
```
### Optional Xet trait
`swift-transformers` includes an `Xet` [package trait](https://github.com/swiftlang/swift-evolution/blob/main/proposals/0450-swiftpm-package-traits.md)
that enables fast, parallel downloads from the Hugging Face Hub
via [swift-xet](https://github.com/huggingface/swift-xet).
Because Xet introduces additional transitive dependencies
(including [AsyncHTTPClient](https://github.com/swift-server/async-http-client)),
it is opt-in.
On Swift 6.1+, enable it in your `Package.swift`:
```swift
dependencies: [
.package(
url: "https://github.com/huggingface/swift-transformers",
from: "1.3.0",
traits: ["Xet"]
)
]
```
Or from the command line:
```bash
swift build --traits Xet
swift test --traits Xet
```
When the trait is not enabled,
Hub downloads use the default `URLSession`-based transport.
On Swift versions earlier than 6.1,
package traits are unavailable and Xet transport cannot be enabled.
> [!NOTE]
> Xcode doesn't yet provide a built-in way to declare package dependencies with traits.
> As a workaround, you can create an internal Swift package
> that re-exports `swift-transformers` with the `Xet` trait enabled,
> then add that package as a local dependency in your Xcode project.
## Projects that use swift-transformers ❤️
- [WhisperKit](https://github.com/argmaxinc/WhisperKit): A Swift Package for state-of-the-art speech-to-text systems from [Argmax](https://github.com/argmaxinc)
- [MLX Swift Examples](https://github.com/ml-explore/mlx-swift-examples): A Swift Package for integrating MLX models in Swift apps.
Using `swift-transformers` in your project? Let us know and we'll add you to the list!
## Other Tools
- [`swift-chat`](https://github.com/huggingface/swift-chat), a simple app demonstrating how to use this package.
- [`exporters`](https://github.com/huggingface/exporters), a Core ML conversion package for transformers models, based on Apple's [`coremltools`](https://github.com/apple/coremltools).
## Contributing
Swift Transformers is a community project and we welcome contributions. Please
check out [Issues](https://github.com/huggingface/swift-transformers/issues)
tagged with `good first issue` if you are looking for a place to start!
Before submitting a pull request, please ensure your code:
- Passes the test suite (`swift test`)
- Passes linting checks (`swift format lint --recursive .`)
To format your code, run `swift format -i --recursive .`.
## License
[Apache 2](LICENSE).