# thinc > A refreshing functional take on deep learning, compatible with your favorite libraries - **URL**: https://www.freshcrate.ai/projects/thinc - **Author**: Explosion - **Category**: RAG & Memory - **Latest version**: `9.1.1` (2026-04-21) - **License**: MIT - **Source**: https://github.com/explosion/thinc - **Language**: Python - **GitHub**: 2,891 stars, 295 forks - **Registry**: pypi (`thinc`) - **Tags**: `pypi` ## Description # Thinc: A refreshing functional take on deep learning, compatible with your favorite libraries ### From the makers of [spaCy](https://spacy.io) and [Prodigy](https://prodi.gy) [Thinc](https://thinc.ai) is a **lightweight deep learning library** that offers an elegant, type-checked, functional-programming API for **composing models**, with support for layers defined in other frameworks such as **PyTorch, TensorFlow and MXNet**. You can use Thinc as an interface layer, a standalone toolkit or a flexible way to develop new models. Previous versions of Thinc have been running quietly in production in thousands of companies, via both [spaCy](https://spacy.io) and [Prodigy](https://prodi.gy). We wrote the new version to let users **compose, configure and deploy custom models** built with their favorite framework. [![tests](https://github.com/explosion/thinc/actions/workflows/tests.yml/badge.svg)](https://github.com/explosion/thinc/actions/workflows/tests.yml) [![Current Release Version](https://img.shields.io/github/v/release/explosion/thinc.svg?include_prereleases&sort=semver&style=flat-square&logo=github)](https://github.com/explosion/thinc/releases) [![PyPi Version](https://img.shields.io/pypi/v/thinc.svg?include_prereleases&sort=semver&style=flat-square&logo=pypi&logoColor=white)](https://pypi.python.org/pypi/thinc) [![conda Version](https://img.shields.io/conda/vn/conda-forge/thinc.svg?style=flat-square&logo=conda-forge&logoColor=white)](https://anaconda.org/conda-forge/thinc) [![Python wheels](https://img.shields.io/badge/wheels-%E2%9C%93-4c1.svg?longCache=true&style=flat-square&logo=python&logoColor=white)](https://github.com/explosion/wheelwright/releases) [![Code style: black](https://img.shields.io/badge/code%20style-black-000000.svg?style=flat-square)](https://github.com/ambv/black) [![Open demo in Colab][colab]][intro_to_thinc_colab] ## 🔥 Features - **Type-check** your model definitions with custom types and [`mypy`](https://mypy.readthedocs.io/en/latest/) plugin. - Wrap **PyTorch**, **TensorFlow** and **MXNet** models for use in your network. - Concise **functional-programming** approach to model definition, using composition rather than inheritance. - Optional custom infix notation via **operator overloading**. - Integrated **config system** to describe trees of objects and hyperparameters. - Choice of **extensible backends**. - **[Read more →](https://thinc.ai/docs)** ## 🚀 Quickstart Thinc is compatible with **Python 3.6+** and runs on **Linux**, **macOS** and **Windows**. The latest releases with binary wheels are available from [pip](https://pypi.python.org/pypi/thinc). Before you install Thinc and its dependencies, make sure that your `pip`, `setuptools` and `wheel` are up to date. For the most recent releases, pip 19.3 or newer is recommended. ```bash pip install -U pip setuptools wheel pip install thinc ``` See the [extended installation docs](https://thinc.ai/docs/install#extended) for details on optional dependencies for different backends and GPU. You might also want to [set up static type checking](https://thinc.ai/docs/install#type-checking) to take advantage of Thinc's type system. > ⚠️ If you have installed PyTorch and you are using Python 3.7+, uninstall the > package `dataclasses` with `pip uninstall dataclasses`, since it may have been > installed by PyTorch and is incompatible with Python 3.7+. ### 📓 Selected examples and notebooks Also see the [`/examples`](examples) directory and [usage documentation](https://thinc.ai/docs) for more examples. Most examples are Jupyter notebooks – to launch them on [Google Colab](https://colab.research.google.com) (with GPU support!) click on the button next to the notebook name. | Notebook | Description | | --------------------------------------------------------------------------------------------------------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | | [`intro_to_thinc`][intro_to_thinc]
[![Open in Colab][colab]][intro_to_thinc_colab] | Everything you need to know to get started. Composing and training a model on the MNIST data, using config files, registering custom functions and wrapping PyTorch, TensorFlow and MXNet models. | | [`transformers_tagger_bert`][transformers_tagger_bert]
[![Open in Colab][colab]][transformers_tagger_bert_colab] | How to use Thinc, `transformers` and PyTorch to train a part- ## Recent releases | Version | Date | Urgency | Changes | | --- | --- | --- | --- | | `9.1.1` | 2026-04-21 | Low | Imported from PyPI (9.1.1) | | `v8.3.13` | 2026-03-23 | Medium | - Add clear error message for float64 dtype mismatch in BLIS gemm (#461). - Document that Model.from_disk requires matching architecture (#751). - Document that Loss.get_grad computes gradient w.r.t. logits, not post-softmax probabilities (#901). - Remove stale comments about torch.cuda.amp.autocast migration (#967). - Fix numpy deprecation warning filter syntax in pyproject.toml. - Ignore NumPy 2.4 align deprecation from pickled dtype in test data. | | `v8.3.13` | 2026-03-23 | Low | - Add clear error message for float64 dtype mismatch in BLIS gemm (#461). - Document that Model.from_disk requires matching architecture (#751). - Document that Loss.get_grad computes gradient w.r.t. logits, not post-softmax probabilities (#901). - Remove stale comments about torch.cuda.amp.autocast migration (#967). - Fix numpy deprecation warning filter syntax in pyproject.toml. - Ignore NumPy 2.4 align deprecation from pickled dtype in test data. | | `v8.3.13` | 2026-03-23 | Low | - Add clear error message for float64 dtype mismatch in BLIS gemm (#461). - Document that Model.from_disk requires matching architecture (#751). - Document that Loss.get_grad computes gradient w.r.t. logits, not post-softmax probabilities (#901). - Remove stale comments about torch.cuda.amp.autocast migration (#967). - Fix numpy deprecation warning filter syntax in pyproject.toml. - Ignore NumPy 2.4 align deprecation from pickled dtype in test data. | | `v8.3.13` | 2026-03-23 | Low | - Add clear error message for float64 dtype mismatch in BLIS gemm (#461). - Document that Model.from_disk requires matching architecture (#751). - Document that Loss.get_grad computes gradient w.r.t. logits, not post-softmax probabilities (#901). - Remove stale comments about torch.cuda.amp.autocast migration (#967). - Fix numpy deprecation warning filter syntax in pyproject.toml. - Ignore NumPy 2.4 align deprecation from pickled dtype in test data. | | `v8.3.13` | 2026-03-23 | Low | - Add clear error message for float64 dtype mismatch in BLIS gemm (#461). - Document that Model.from_disk requires matching architecture (#751). - Document that Loss.get_grad computes gradient w.r.t. logits, not post-softmax probabilities (#901). - Remove stale comments about torch.cuda.amp.autocast migration (#967). - Fix numpy deprecation warning filter syntax in pyproject.toml. - Ignore NumPy 2.4 align deprecation from pickled dtype in test data. | | `v8.3.13` | 2026-03-23 | Low | - Add clear error message for float64 dtype mismatch in BLIS gemm (#461). - Document that Model.from_disk requires matching architecture (#751). - Document that Loss.get_grad computes gradient w.r.t. logits, not post-softmax probabilities (#901). - Remove stale comments about torch.cuda.amp.autocast migration (#967). - Fix numpy deprecation warning filter syntax in pyproject.toml. - Ignore NumPy 2.4 align deprecation from pickled dtype in test data. | | `v8.3.13` | 2026-03-23 | Low | - Add clear error message for float64 dtype mismatch in BLIS gemm (#461). - Document that Model.from_disk requires matching architecture (#751). - Document that Loss.get_grad computes gradient w.r.t. logits, not post-softmax probabilities (#901). - Remove stale comments about torch.cuda.amp.autocast migration (#967). - Fix numpy deprecation warning filter syntax in pyproject.toml. - Ignore NumPy 2.4 align deprecation from pickled dtype in test data. | | `v8.3.13` | 2026-03-23 | Low | - Add clear error message for float64 dtype mismatch in BLIS gemm (#461). - Document that Model.from_disk requires matching architecture (#751). - Document that Loss.get_grad computes gradient w.r.t. logits, not post-softmax probabilities (#901). - Remove stale comments about torch.cuda.amp.autocast migration (#967). - Fix numpy deprecation warning filter syntax in pyproject.toml. - Ignore NumPy 2.4 align deprecation from pickled dtype in test data. | | `v8.3.13` | 2026-03-23 | Low | - Add clear error message for float64 dtype mismatch in BLIS gemm (#461). - Document that Model.from_disk requires matching architecture (#751). - Document that Loss.get_grad computes gradient w.r.t. logits, not post-softmax probabilities (#901). - Remove stale comments about torch.cuda.amp.autocast migration (#967). - Fix numpy deprecation warning filter syntax in pyproject.toml. - Ignore NumPy 2.4 align deprecation from pickled dtype in test data. | ## Citation - HTML: https://www.freshcrate.ai/projects/thinc - Markdown: https://www.freshcrate.ai/projects/thinc.md - Dependencies JSON: https://www.freshcrate.ai/api/projects/thinc/deps _Generated by freshcrate.ai. Indexes pypi releases for AI-agent ecosystem packages._