Who contributed to Hugging Face on June 29, 2026?

2 developers shipped this update, including dxqb and Cyrilvallez.

What were the notable Hugging Face updates?

Fix Kohya UNet LoRA key conversion for conv_in/conv_out/time_embedding (#14006), Fix Kohya UNet LoRA key conversion for conv_in/conv_out/time_embedding, and Fix auto-mappings registration for remote code & fixes a few custom code issues.

@huggingface

Transformers, Datasets, and the open AI-model layer

github ↗

Pick a date

Topics: Python AI / ML Full archive →

The Wire · Showcase

DIFFUSERS FIXES KOHYA LORA COMPATIBILITY BUG BLOCKING MODEL LOADING

By RepoJournal · Filed 06:03 UTC on June 29, 2026 · About Hugging Face

2 people shipped this

dxqb @dxqb 1 cited

Cyrilvallez @Cyrilvallez 1 cited

Kohya-format LoRA weights for conv_in, conv_out, and time_embedding layers were silently failing to load in diffusers, now fixed.

The diffusers library was dropping Kohya-format UNet LoRA keys on three critical top-level submodules [1], causing them to be reported as unexpected rather than applied to the model. This affected anyone fine-tuning these layers with Kohya-ss tooling and trying to load checkpoints into diffusers. The fix adds proper key conversion mapping so lora_unet_conv_in and similar patterns now correctly map to diffusers naming conventions [2]. Meanwhile on the transformers side, auto-mappings registration for remote code is getting fixed [3] to handle model loading edge cases that surfaced in earlier patches. The transformers team also hardened device pinning in GDN mixer layers [4], preventing unexpected cuda:0 allocation when target devices were specified elsewhere. Swift-transformers added Sendable conformance to Hub.Repo and Hub.RepoType [5], unblocking Swift 6 strict concurrency checks for anyone caching or passing these types across task boundaries.

Action items

→ If using Kohya LoRA fine-tunes with diffusers, pull the latest version to restore conv_in/conv_out/time_embedding loading huggingface/diffusers [immediate]
→ Review transformers remote code loading paths if you're using custom model implementations huggingface/transformers [plan]
→ Swift developers: update swift-transformers to use Repo in concurrent contexts without warnings huggingface/swift-transformers [plan]

References

[1] Fix Kohya UNet LoRA key conversion for conv_in/conv_out/time_embedding (#14006) huggingface/diffusers
[2] Fix Kohya UNet LoRA key conversion for conv_in/conv_out/time_embedding ↗ huggingface/diffusers
[3] Fix auto-mappings registration for remote code & fixes a few custom code issues ↗ huggingface/transformers
[4] Don't pin the gated delta net norm to `cuda:0` with a hardcoded device (#46817) huggingface/transformers
[5] Add Sendable conformance to Hub.Repo and Hub.RepoType (#374) huggingface/swift-transformers

Quick answers

What shipped in Hugging Face on June 29, 2026?: Kohya-format LoRA weights for conv_in, conv_out, and time_embedding layers were silently failing to load in diffusers, now fixed. In total, 7 commits and 8 pull requests landed.
Who contributed to Hugging Face on June 29, 2026?: 2 developers shipped this update, including dxqb and Cyrilvallez.
What were the notable Hugging Face updates?: Fix Kohya UNet LoRA key conversion for conv_in/conv_out/time_embedding (#14006), Fix Kohya UNet LoRA key conversion for conv_in/conv_out/time_embedding, and Fix auto-mappings registration for remote code & fixes a few custom code issues.

TRANSFORMERS-CI SHIPS INTELLIGENT TEST TRIAGE, CUTS LLM NOISE 100X

Hugging Face's serge verify loop now distills 100KB failure tracebacks down to the actual assertion, runs every test 5 times to catch flakiness, and dispatches each failure category with a custom instruction before spinning GPU or consuming a single LLM token.

python 45 shipped 2-min read

@huggingface 2 days ago

TRL CI BURNS, BITSANDBYTES PINNED BELOW 0.50.0

Hugging Face's TRL training library is firefighting two separate CI failures overnight, forcing temporary pins on bitsandbytes and xfailing critical tests against a broken transformers dev build.

python 7 shipped 1-min read

@huggingface 3 days ago

TRANSFORMERS OVERHAULS LINEAR ATTENTION WHILE DEPRECATING LEGACY RESPONSE SCHEMA

The transformers library is retiring its fragile response_schema prototype in favor of streaming-compatible parsing, while simultaneously refactoring every linear attention model to use standardized convolution patterns.

python 70 shipped 2-min read

@huggingface 4 days ago

TRANSFORMERS SHIPS FSDP DISTRIBUTED TRAINING STACK, HUB LIBRARY PLUGS REDOS HOLE

Hugging Face landed distributed training orchestration in transformers while plugging a regex vulnerability that could stall untrusted card parsing for minutes.

+10

python 61 shipped 1-min read

Elsewhere on the wire

AI Agents about 21 hours ago

CLINE SHIPS FIVE CRITICAL AGENT FIXES AFTER END-TO-END TESTING BLITZ

Saoudrizwan just landed a cascade of fixes that prevent the agent loop from running wild, stop file reads from resolving to the wrong directory, and plug the compaction pipeline.

ai-agents 16 shipped 2-min read

Local LLMs about 21 hours ago

LLAMA.CPP SHIPS MINIMAX-M3 SPARSE ATTENTION, VLLM UNLOCKS HETEROGENEOUS PREFILL-DECODE

The inference stack gets a new model architecture and a fundamental shift in how prefill and decode can run on different hardware configurations.

+10

llms 125 shipped 1-min read

@CachyOS about 21 hours ago

GLIBC 2.43 REBUILD SHIPS, INSTALLER JUMPS TO C++26

CachyOS pushed glibc 2.43 through the pipeline while the CLI installer upgraded its C++ standard to C++26, signaling a major refresh across core infrastructure.

infra 97 shipped 1-min read

Elixir & Phoenix about 21 hours ago

ECTO FIXES SUBQUERY PLANNING IN JOINS, ELIXIR PATCHES IEX PARSING

Ecto's query planner now handles subqueries correctly in interpolated join filters, while Elixir fixes a syntax error that broke 'not in' as a continuation operator in IEx.