# Hugging Face Transformers v4.37.0 — Release & Upgrade Intelligence
- **Product:** Hugging Face Transformers (https://huggingface.co)
- **Version:** v4.37.0
- **Released:** Jan 22, 2024
- **Impact Score:** 5/10
- **Breaking Changes:** Not verified
- **Canonical Analysis:** https://changelogs.us/transformers/updates/v4.37.0/

## Executive Summary (TL;DR)
Hugging Face Transformers v4.37.0. Source highlight: Qwen2 is the new model series of large language models from the Qwen team. Previously, the Qwen series was released, including Qwen-72B, Qwen-1.8B, Qwen-VL, Qwen-Audio, etc.

## New Features
- Add qwen2  by @JustinLin610 in #28436
- [Phi2] Add support for phi2 models  by @susnato in #28211
- update docs to add the `phi-2` example  by @susnato in #28392
- Add SigLIP  by @NielsRogge in #26522
- Adds VIP-llava to transformers  by @younesbelkada in #27932
- Add FastSpeech2Conformer  by @connor-henderson in #23439

## Improvements
- [`Awq`] Enable the possibility to skip quantization for some target modules  by @younesbelkada in #27950
- [docs] Fused AWQ modules  by @stevhliu in #27896
- [docs] Sort es/toctree.yml | Translate performance.md (#28262)
- [DETA] Improvement and Sync from DETA especially for training  by @SangbumChoi in #27990

## Bug Fixes
- Fixes default value of `softmax_scale` in `PhiFlashAttention2`.  by @gugarosa in #28537
- Fix Vip-llava docs  by @younesbelkada in #28085
- Fix SDPA correctness following torch==2.1.2 regression  by @fxmarty in #27973
- Fix SDPA tests  by @fxmarty in #28552
- Generate: fix speculative decoding  by @gante in #28166
- Fix _speculative_sampling implementation  by @ofirzaf in #28508

## Upgrade Command
```bash
python -m pip install 'transformers==4.37.0'
```