●Change8

Migrating to Transformers v5.14.0

Version v5.14.0 introduces 2 breaking changes. This guide details how to update your code.

Released: 7/15/2026

2
Breaking Changes
2
Migration Steps
18
Affected Symbols

⚠️ Check Your Code

If you use any of these symbols, you need to read this guide:

GPTNeoXGPTBigCodeLocalLayerRepositoryDeepGEMMSDPAFlashAttentionStaticCacheQwen3-VLDETR-family sine position embeddingssiglipQwen OmniAqlmConfigDeepSeek R1DtensorExecutorchremap_legacy_layer_typesDistributedConfigmlinter

Breaking Changes

●Issue #1

GPTNeoX now remaps `embed_out` to `lm_head`. Users relying on the previous weight naming should update their code.

●Issue #2

GPTBigCode has `_supports_attention_backend = True` enabled for vLLM compatibility. Users relying on the previous attention backend behavior should update their code.

Migration Steps

  1. 1
    Update code to use `lm_head` instead of `embed_out` for GPTNeoX.
  2. 2
    Ensure compatibility with `_supports_attention_backend = True` for GPTBigCode if relying on previous attention backend behavior.

Release Summary

This release introduces the Inkling multimodal model and TIPSv2, alongside significant improvements in kernel performance, generation capabilities, and bug fixes across various components. It also includes breaking changes for GPTNeoX and GPTBigCode related to weight remapping and attention backend compatibility.

Need More Details?

View the full release notes and all changes for Transformers v5.14.0.

View Full Changelog