# Hugging Face Transformers v4.50.0 — Release & Upgrade Intelligence
- **Product:** Hugging Face Transformers (https://huggingface.co)
- **Version:** v4.50.0
- **Released:** Mar 21, 2025
- **Impact Score:** 7/10
- **Breaking Changes:** Source flagged
- **Canonical Analysis:** https://changelogs.us/transformers/updates/v4.50.0/

## Executive Summary (TL;DR)
Hugging Face Transformers v4.50.0. Source highlight: Model-based releases

## Breaking Changes & Deprecations
- Fixing the docs corresponding to the breaking change in torch 2.6.  by @Narsil in #36420

## New Features
- Add aya  by @ArthurZucker in #36521
- Add Mistral3  by @Cyrilvallez in #36790
- Add SigLIP 2  by @qubvel in #36323
- Add Prompt Depth Anything Model by @haotongl in #35401
- Add attention visualization tool by @ArthurZucker in #36630
- Added Support for Custom Quantization  by @keetrap in #35915

## Improvements
- chore: enhance messages in docstrings (#36525)
- chore: enhance message descriptions in parameters,comments,logs and docstrings (#36554)
- Use pyupgrade --py39-plus to improve code (#36843)

## Bug Fixes
- [tests] fix `EsmModelIntegrationTest::test_inference_bitsandbytes`   by @faaany in #36225
- Fix `LlavaForConditionalGenerationModelTest::test_config` after #36077  by @ydshieh in #36230
- AMD DeepSpeed image additional HIP dependencies  by @ivarflakstad in #36195
- [generate] remove cache v4.47 deprecations  by @gante in #36212
- Add missing atol to torch.testing.assert_close where rtol is specified  by @ivarflakstad in #36234
- [tests] remove tf/flax tests in `/generation`  by @gante in #36235

## Upgrade Command
```bash
python -m pip install 'transformers==4.50.0'
```