Uform

[2.1.1](https://github.com/unum-cloud/uform/compare/v2.1.0...v2.1.1) (2024-04-16)


Fix

* Importing ViT in `gen_model.py` (80) ([21f49ba](https://github.com/unum-cloud/uform/commit/21f49bab444e0761ab7fc7ed20b3c81fb7924d17)), closes [#80](https://github.com/unum-cloud/uform/issues/80)

[2.1.0](https://github.com/unum-cloud/uform/compare/v2.0.2...v2.1.0) (2024-04-14)


Add

* Initial Swift support ([00bd84c](https://github.com/unum-cloud/uform/commit/00bd84c59995c7b3daa0b4fa1597f77608806fdb))

Fix

* Image preprocessing in Swift ([f2772d0](https://github.com/unum-cloud/uform/commit/f2772d0d92317818c4d1c49166bc7ec3ee314f60))

Improve

* Fetching nested configs ([729b9d9](https://github.com/unum-cloud/uform/commit/729b9d9f73990f2689c22af593171184589a2b27))

Make

* Formatting Swift code ([f6faf4c](https://github.com/unum-cloud/uform/commit/f6faf4cd877f6034d8c66edb108bc07ec1735232))

[2.0.2](https://github.com/unum-cloud/uform/compare/v2.0.1...v2.0.2) (2024-03-28)


Make

* Fix PyPi CI version with hash ([364afe6](https://github.com/unum-cloud/uform/commit/364afe605540c5ca8649c6b798ce4d8e65540c7c))

[2.0.1](https://github.com/unum-cloud/uform/compare/v2.0.0...v2.0.1) (2024-03-28)


Make

* PyPi upload version ([9453802](https://github.com/unum-cloud/uform/commit/945380263c5a8d6bc7906afedd7728ed1ad0cfcb))


Today we are releasing a new batch of multimodal models trained with [Nebius](https://nebius.ai) and already available on [HuggingFace](https://huggingface.co/unum-cloud) 🤗

1. Matryoshka style multimodal embeddings ranging from 64 to 256 and 768 dimensions 🖼️
2. Improved multimodal chat in 1.2B parameters, tuned with Direct Preference Optimization 💬
3. ONNX backend, making PyTorch dependency optional for lightning fast deployments ⚡

Great thanks to lmmx, blackforestboi, and kapulkin for their patches to the project!

---

* Performance observations for M2 CPUs (56) ([8374ef6](https://github.com/unum-cloud/uform/commit/8374ef6a4c13ec6875a6a349aa5297ceee47d6d3)), closes [#56](https://github.com/unum-cloud/uform/issues/56)
* Passing labels to `text_decoder` to compute loss. (65) ([f445a8b](https://github.com/unum-cloud/uform/commit/f445a8b73faa8fd6a83b18cc547660d45eebfd5a)), closes [#65](https://github.com/unum-cloud/uform/issues/65)
* Larger batch benchmarks ([fdc8587](https://github.com/unum-cloud/uform/commit/fdc85876162fa631715dd1643f0ffe53f92a04e2))
* pre-commit config and linters (62) ([0a3efac](https://github.com/unum-cloud/uform/commit/0a3efac1f6b80295d14b2cde291b7b7f20a82284)), closes [#62](https://github.com/unum-cloud/uform/issues/62)