# Hugging Face Transformers v4.46.0 — Release & Upgrade Intelligence
- **Product:** Hugging Face Transformers (https://huggingface.co)
- **Version:** v4.46.0
- **Released:** Oct 24, 2024
- **Impact Score:** 5/10
- **Breaking Changes:** Not verified
- **Canonical Analysis:** https://changelogs.us/transformers/updates/v4.46.0/

## Executive Summary (TL;DR)
Hugging Face Transformers v4.46.0. Source highlight: The Moshi model was proposed in Moshi: a speech-text foundation model for real-time dialogue by Alexandre Défossez,

## New Features
- Add Zamba  by @pglorio in #30950
- add Glm  by @Cyrilvallez in #33823
- Add Idefics 3!  by @andimarafioti in #32473
- Add SynthID (watermerking by Google DeepMind) by @gante in #34350
- FEAT : Adding BitNet quantization method to HFQuantizer  by @MekkCyber in #33410
- Add gguf support for bloom  by @VladOS95-cyber in #33473

## Bug Fixes
- Fix Gradient Accumulation issue by @ArthurZucker in #34191
- Enable users to use their own loss functions + deal with prefetching for grad accum  by @muellerzr in #34198
- Enable Gradient Accumulation fix across all models + trainer fully in forward() by @muellerzr #34283
- adding positional encoder changes and tests  by @manuelsh in #32600
- Uniformize kwargs for chameleon processor  by @leloykun in #32181
- [`MllamaProcessor`] Update errors and API with multiple image  by @ArthurZucker in #33715

## Upgrade Command
```bash
python -m pip install 'transformers==4.46.0'
```