← back to report

huggingface/transformers

// all events — 2026-10-07

Type AI Score Description Actor Reason Date
commit 1.00 Fix DataCollatorForLanguageModeling ignoring seed=0 (#49349) Commit message contains explicit AI assi 2026-10-06
commit 1.00 GPT-OSS: read the clamped SwiGLU's alpha and limit from the Commit message contains explicit AI assi 2026-10-06
commit 1.00 Bump doc-builder main docs workflow pin (#49350) Commit message contains explicit AI assi 2026-10-06
commit 1.00 Fix run_glue.py label ids becoming strings with pandas>=3 (# Commit message contains explicit AI assi 2026-10-05
commit 1.00 Fix chunked prefill with multi-axis position ids (Qwen3-VL) Commit message contains explicit AI assi 2026-10-05
commit 1.00 Do RoPE with mul instead of matmul (#49265) Commit message contains explicit AI assi 2026-10-05
commit 1.00 Enable model test on upcoming `torch_tpu` backend (#49207) Commit message contains explicit AI assi 2026-10-05
commit 1.00 [CI] Avoid fetching all artifacts when only specific ones ar Commit message contains explicit AI assi 2026-10-03
commit 1.00 [CI] check_bad_commit: use EFS cache to avoid Xet FUSE OOM ( Commit message contains explicit AI assi 2026-10-02
commit 1.00 [ESM/Persimmon/NllbMoe/ESMFold] use safetensors repos to fix Commit message contains explicit AI assi 2026-10-01
commit 1.00 QA: restore masking_utils export comment with a noqa (#49209 Commit message contains explicit AI assi 2026-09-30
commit 1.00 Replace datasets that no longer load with maintained uploads Commit message contains explicit AI assi 2026-09-30
commit 1.00 Finish removing the MPS autocast workaround (#49157) Commit message contains explicit AI assi 2026-09-29
commit 1.00 Document running the example scripts on Hugging Face Jobs (# Commit message contains explicit AI assi 2026-09-29
commit 1.00 [CacheHardIntegrationTest] use dedicated safetensors repo to Commit message contains explicit AI assi 2026-09-29
commit 1.00 [imagegpt/vilt/trocr] fix cache fixture tests: use hf_hub_do Commit message contains explicit AI assi 2026-09-29
pr 1.00 [Docker] Optimize FA2/FA3/FA4 install in transformers-all-la Explicitly states it was generated with 2026-10-06
pr 1.00 Complete PR #49228: remove unused device=None from rope __in Explicitly states it was generated with 2026-10-03
pr 1.00 Link the padding side section from SSM and hybrid model docs PR body explicitly mentions AI collabora 2026-10-02
pr 1.00 Generation cleanup in pipelines Explicitly states that the PR is mostly 2026-10-02
pr 1.00 Composite transparent images in shared RGB conversion and lo Explicitly states that AI assistance was 2026-09-24
pr 1.00 Fix Phi3 LongRoPE cache rebuild at the context boundary The description explicitly states that A 2026-10-06
pr 1.00 Resolve negative `output_hidden_states` layer indices PR body explicitly mentions AI collabora 2026-10-06
pr 1.00 Upgrade Python Min version to `3.11` PR body explicitly mentions AI collabora 2026-10-02
pr 1.00 Fix DataCollatorForLanguageModeling ignoring seed=0 PR body explicitly mentions AI collabora 2026-10-06
pr 1.00 fix(cohere_compass): add video to input_modalities (#49320) PR body explicitly mentions AI collabora 2026-10-05
pr 1.00 fix(dsa): Chunk the lightning indexer scores over the query The description explicitly states that A 2026-10-05
pr 1.00 GPT-OSS: read the clamped SwiGLU's alpha and limit from the PR body explicitly mentions AI collabora 2026-10-06
pr 1.00 Use Siglip2Tokenizer when the Hub tokenizer_class is GemmaTo PR body explicitly mentions AI collabora 2026-10-02
pr 1.00 Remove the GitHub API cache-bust from the AMD docker images Explicitly states it was generated with 2026-10-06
pr 1.00 Bump doc-builder main docs workflow pin Explicitly states it was generated with 2026-10-06
pr 1.00 Bump doc-builder workflow pin Explicitly states it was generated with 2026-10-06
pr 1.00 Load deep-gemm on Blackwell: skip the kernels arch check for PR body explicitly mentions AI collabora 2026-10-06
pr 1.00 Replicate 0-dim parameters in TP/EP sharding PR body explicitly mentions AI collabora 2026-10-06
pr 1.00 Fix WatermarkDetector repeated-ngram counting PR body explicitly mentions AI collabora 2026-10-05
pr 1.00 Warn on pre_tokenizer mismatch between tokenizer.json and to PR body explicitly mentions AI collabora 2026-10-06
pr 1.00 Finegrained kernels: unified quantization kernels api (separ PR body explicitly mentions AI collabora 2026-09-22
pr 1.00 [Generation] Fix WatermarkDetector batch ordering bug with m PR body explicitly mentions AI collabora 2026-10-05
pr 1.00 fix(modeling_utils): propagate attention implementation to n PR body explicitly mentions AI collabora 2026-10-05
pr 0.35 [security] Add SSRF validation to load_image URL fetch Formal security-report structure is some 2026-07-10
pr 0.28 Add qwen3 tts Polished feature summary is somewhat gen 2026-03-07
pr 0.14 [tests] Fix flaky CMVN tests in Speech2TextFeatureExtraction Specific failure rates and test details 2026-10-06
pr 0.12 Fix Trainer step accounting when GradScaler skips an optimiz Detailed domain-specific explanation wit 2026-10-01
pr 0.12 [tests] Skip onnx.exactness for BitForImageClassification (O Concrete numerical debugging details sug 2026-10-06
pr 0.12 Fix create_dummy_models CI failures for new models Detailed domain-specific explanation, bu 2026-10-05
pr 0.10 Qualcomm AI Engine Direct - Support QNN quantization and low Technical and repository-specific wordin 2026-09-29
pr 0.10 Store sliding layers of the static cache at full size on Neu Technical implementation details and tar 2026-10-06
pr 0.10 Add BerryLM Very polished, but unusually specific ar 2026-09-17
pr 0.08 [Cache] Linear attn cache: use new getters Concise, specific cache behavior discuss 2026-10-06
pr 0.08 fix(tokenizers): prefer `vocab_file` over `tokenizer.json` v Polished but highly specific bug analysi 2026-10-05
pr 0.08 v1/systemone endpoint Mostly human-like with minor polished bo 2026-10-05
pr 0.06 Fix explicit kernelization modes for evaluation models Detailed API and validation references a 2026-10-06
pr 0.05 Add an FP8 backend to `QuantizedCache` Detailed implementation tradeoffs and be 2026-09-10
pr 0.05 [`distributed`] Add dense and expert device mesh Specific distributed-systems details and 2026-09-16
pr 0.04 Adding Omnilingual ASR models Informal progress notes and model-specif 2026-01-13
pr 0.04 Fix Pixtral Flash Attention image boundaries after axial RoP Detailed reproduction, version reference 2026-10-06
pr 0.04 Loading: convert transposed and RoPE-permuted weights before Deep implementation detail, concrete fai 2026-09-04
pr 0.04 [Nemotron3Diarization] Support QK norm and partial rotary in Domain-specific architecture and numeric 2026-10-03
pr 0.04 Allow registering new processing backends Casual abbreviations, typos, and specifi 2026-09-21
pr 0.04 [DON'T MERGE YET] [Qwen3VLMoe] Update expected text in test_ Reproducible commit, runner, and tensor 2026-10-04
pr 0.04 FSDP x TP Saving / Loading Contains specific implementation details 2026-09-14
commit 0.03 [`distributed`] Add expert-parallel token dispatch, default Domain-specific distributed-systems deta 2026-10-06
commit 0.03 [`distributed`] decoupled tp_plan and ep_plan (#48859) Specific implementation details, typos, 2026-10-06
commit 0.03 [`distributed`] Add dense and expert device mesh (#48857) Technical mesh terminology and informal 2026-10-06
commit 0.03 🚨 Return router logits from every MoE model that computes th Highly domain-specific changelog with it 2026-10-05
commit 0.03 Add Trainer.loss_is_scaled_for_ga to declare whether compute Technical, repository-specific wording a 2026-10-02
pr 0.03 handle transpose during model loading with TP (Qwen3-VL-MoE) Concrete tensor-shape reasoning and visi 2026-08-27
pr 0.03 fix nested set_attn_implementation Terse issue references and informal word 2026-09-27
pr 0.03 Fix encoder repetition penalty after score expansion Specific implementation details and envi 2026-10-04
pr 0.03 [CB] Scale the per-token memory estimate by the TP size Highly specific implementation details a 2026-09-06
pr 0.03 [Umbrella] Continuous batching fixes for training-while-gene Contains concrete benchmarks, issue refe 2026-08-25
pr 0.03 fix(gemma): default GemmaTokenizer add_bos_token to True Polished but highly specific technical e 2026-10-06
pr 0.03 fix(pipelines): accept code_revision via model_kwargs withou Detailed repository-specific diagnosis i 2026-10-06
pr 0.03 Fix data-dependent branch in _ignore_causal_mask_sdpa under Highly specific compiler and generation 2026-10-06
pr 0.03 [`distributed`] Add expert-parallel token dispatch, default Detailed domain-specific explanation inc 2026-09-16
pr 0.03 [`distributed`] decoupled tp_plan and ep_plan Specific distributed-systems explanation 2026-09-16
pr 0.03 FSDP x TP Saving / Loading, Optimizer Concise, repository-specific technical w 2026-10-02
pr 0.03 remove colwise_gather_output for lm_head for DeepseekV4 Informal technical explanation with typo 2026-10-05
commit 0.02 Smthn smthn (#49364) Highly specific implementation details, 2026-10-06
commit 0.02 Move CI to Python 3.11, with the version set in `.python-ver Technical CI details and terse commit hi 2026-10-06
commit 0.02 Allow registering new processing backends (#48984) Informal terse notes and concrete debugg 2026-10-06
commit 0.02 remove colwise_gather_output for lm_head for DeepseekV4 (#4 Terse, repository-specific revert notes 2026-10-06
commit 0.02 Fix quantized cache (#48700) Detailed implementation and test notes a 2026-10-05
commit 0.02 🚨 Select the OWLv2 image query by objectness (#49200) Specific technical change with a normal 2026-10-02
commit 0.02 Support per-layer cache configuration (#48178) Long iterative history contains specific 2026-10-02
commit 0.02 Remove rotaries deprecation cycle (#49228) Terse, informal commit-style wording sug 2026-10-02
commit 0.02 [CB] Make CB more device agnostic and add support for XPU (# Detailed domain-specific summary and ite 2026-10-01
pr 0.02 Let token dispatch run with `ep_size != tp_size` Concise, issue-specific technical explan 2026-10-06
pr 0.02 🚨🚧 FeatureExtractor → AudioProcessor Minimal informal free text is consistent 2026-03-02
pr 0.02 Simplify group images by shape Concise project-specific bullets and tes 2026-08-21
pr 0.02 Copy default response parser args Specific bug context and casual phrasing 2026-10-06
pr 0.02 [CB] Varlen path: copy only the used prefix of the index sto Specific benchmarks, hardware, and repro 2026-09-06
pr 0.02 Tensor parallelism: let a split model be one replica among s Highly specific distributed-systems deta 2026-09-06
pr 0.02 Remove deprecated data helpers Informal phrasing, typos, and specific r 2026-10-06
pr 0.02 🚨 Deprecate low use models Informal wording, typos, and a concrete 2026-09-25
pr 0.02 [WIP] modular cleanup Specific technical plan, dates, and info 2026-10-02
pr 0.02 [`distributed`] Default MoE `ep_plans` to token dispatch Brief technical summary with typos and r 2026-09-28
pr 0.02 Fix minicpm 4.6V video batching and CI Informal, typo-filled wording and specif 2026-10-02
pr 0.02 Validate Gemma4 audio placeholders in processor inputs Detailed reproducer and test specifics a 2026-07-04
commit 0.01 Let token dispatch run with `ep_size != tp_size` (#49368) Terse technical title and standard human 2026-10-07
commit 0.01 [tests] Skip onnx.exactness for BitForImageClassification (O Specific ORT behavior, tolerance, and mo 2026-10-06
commit 0.01 Link the padding side section from SSM and hybrid model docs Concrete documentation references and it 2026-10-06
commit 0.01 Revert "[CI] Temporarily disable AMD scheduled CI trigger" ( Short conventional revert description wi 2026-10-06
commit 0.01 Fix explicit kernelization modes for evaluation models (#493 Brief technical fix with review iteratio 2026-10-06
commit 0.01 fix(cohere_compass): add video to input_modalities (#49320) Conventional commit syntax and precise m 2026-10-06
commit 0.01 [`distributed`] Default MoE `ep_plans` to token dispatch (#4 Detailed domain-specific history and ter 2026-10-06
commit 0.01 [`distributed`]: adapt Trainer to work Expert Parallel (#488 Specific distributed-training details an 2026-10-06
commit 0.01 Bump doc-builder workflow pin (#49342) Short, specific dependency-pin updates a 2026-10-06
commit 0.01 Fix WatermarkDetector repeated-ngram counting (#49315) Specific regression and review reference 2026-10-06
commit 0.01 Fix minicpm 4.6V video batching and CI (#49259) Very terse informal debugging notes stro 2026-10-06
commit 0.01 Fix the `labels` docstring of CLIPSegForImageSegmentation (# Concise, technically specific documentat 2026-10-05
commit 0.01 [CI] Temporarily disable AMD scheduled CI trigger (#49271) Brief CI change with a normal co-author 2026-10-05
commit 0.01 Remove the unused pandas entry from setup.py deps (#49329) Specific dependency and CI-version detai 2026-10-05
commit 0.01 [Fix] Remove old test file with two ancient tests (#49280) Short, informal commit sequence typical 2026-10-05
commit 0.01 Fix multi-image feature ordering during generation batch exp Concise fix description with a normal co 2026-10-05
commit 0.01 Why override when you can fix in `Mixin` (#49105) Fragmentary informal commit notes strong 2026-10-05
commit 0.01 Deprecated pipelines should redirect; removed pipelines shou Brief review-driven changelog with a nor 2026-10-05
commit 0.01 Fix encoder repetition penalty source rows after batch expan Terse, domain-specific commit message wi 2026-10-05
commit 0.01 Preserve masked EOS scores in exponential decay length penal Concise technical commit message; no sig 2026-10-05
commit 0.01 Revert "[CI] ssh-runner: add optional cache_type input to sw Concise revert message with a specific c 2026-10-02
commit 0.01 Don't forward text-level flash-attention kwargs to multimoda Terse commit notes and an informal CI re 2026-10-02
commit 0.01 [`GTE`] Add memory mixin (#49234) Terse, informal wording is characteristi 2026-10-01
commit 0.01 [docs] TP for PEFT adapters (#49060) Brief, informal commit notes lack AI-sty 2026-10-01
commit 0.01 correct final metrics due to resuming from checkpoint (#4920 Concise wording is typical of a human co 2026-10-01
commit 0.01 🚨 Attention 🚨 Deprecate "paged|" prefix for SDPA and flash Terse commit sequence and domain-specifi 2026-10-01
commit 0.01 🚨 Attention 🚨 Make regular attention support CB (#49101) Informal review notes, typos, and co-aut 2026-10-01
commit 0.01 🚨 [CB] 🚨 Fuse update for index and block table path (#49088) Minimal phrase is a typical human squash 2026-10-01
commit 0.01 [Refactor] Make some flash-attention utils more readable (#4 Short technical change list and informal 2026-10-01
commit 0.01 🚨 [CB] 🚨 Little fixes before removing "paged" (#49069) Terse technical notes and shorthand are 2026-10-01
commit 0.01 fix the slack API call (#49220) Simple wording and a specific workflow f 2026-10-01
pr 0.01 Move processing defaults to class Informal technical wording and reviewer 2026-09-24
pr 0.01 Byte-level tokenizer.json + Hub tokenizer_class LlamaTokeniz Brief, informal wording and grammatical 2026-08-25
pr 0.01 Add MOSS-Transcribe-Diarize model Extremely terse description with an info 2026-08-28
pr 0.01 [Umbrella] FSDP2 + expert-parallel token dispatch, with shar Terse, repository-specific summary with 2026-08-22
pr 0.01 Fix offloaded cache prefetch racing with the offloading copy Concise, concrete debugging report with 2026-10-06
pr 0.01 🚨🚨🚨 luke fixe Typos, informal phrasing, and detailed t 2026-08-29
pr 0.01 Remove examples Informal typos and personal review remar 2026-10-06
pr 0.01 :rotating_light: SubConfigSpec for centralized init of subco Contains typos, informal phrasing, and d 2026-08-25
pr 0.01 More RoPE simplification, TF32 fixes Informal phrasing, jargon, and emoji str 2026-10-05
pr 0.01 Fix static-cache generate() with host-side inputs and under Concrete tests, implementation details, 2026-10-05
pr 0.01 Move CI to Python 3.11, with the version set in `.python-ver Brief, concrete CI change summary is con 2026-10-05
pr 0.01 Trainer: `save_pretrained` on every rank for PEFT model Very terse description with a natural ty 2026-10-05
pr 0.01 Fix padding mask skipped for batch size 1 in linear-attentio Highly specific debugging details and be 2026-06-19
commit 0.00 ALM testx fIxing... (#49355) Fragmentary typo-filled message is stron 2026-10-06
commit 0.00 Fix pp chart2table inference (#49225) Informal typo-filled notes are character 2026-10-02
commit 0.00 fix precomputed flash kwargs clashing (#47928) Extremely terse and informal commit mess 2026-10-02
commit 0.00 Reapply modular examples (#49230) Brief, informal commit notes with no AI- 2026-10-02
commit 0.00 Remove some stuff that does not exist (#49232) Terse wording and informal phrasing indi 2026-10-02
commit 0.00 Remove deprecation cycle on mask and cache (#49231) Minimal, informal revert description. 2026-10-02
commit 0.00 Dev update (#49212) — 2026-09-30
commit 0.00 model: Add GTE to Transformers (#48416) — 2026-09-30
commit 0.00 [Nemotron3Diarization] fix streaming last stft frame dropped — 2026-09-30
commit 0.00 Fix missing router_logits in Qwen3.5-MoE and other MoE model — 2026-09-30
commit 0.00 Fix MiniMax M3 partial 3D vision rotary embeddings (#49164) — 2026-09-30
commit 0.00 Add Strix Halo (gfx1151) Atlas Inference Hub-kernel path for — 2026-09-30
commit 0.00 Fix MPS GQA version gating (#49210) — 2026-09-30
commit 0.00 Fix additional_special_tokens data loss with extra_special_t — 2026-09-30
commit 0.00 Fix stale _added_tokens_encoder entries in cpmant and wav2ve — 2026-09-30
commit 0.00 Fix deepstack features for mixed-input (#49177) — 2026-09-30
commit 0.00 gguf user defined tokens (#49008) — 2026-09-30
commit 0.00 [`Kernels`] Sync mamba version (#49205) — 2026-09-30
commit 0.00 Map bare list, tuple and dict annotations to the right JSON — 2026-09-30
commit 0.00 Fix SequenceBiasLogitsProcessor edge cases: token id 0 and p — 2026-09-30
commit 0.00 Fail fast on eval OOM under `auto_find_batch_size` (#49198) — 2026-09-30
commit 0.00 [MPS] Let mps sdpa handle grouped query attention directly ( — 2026-09-30
commit 0.00 [Parakeet] Convert NeMo's stochastic depth to layerdrop (#49 — 2026-09-30
commit 0.00 Video processors - general maintenance (#48251) — 2026-09-30
commit 0.00 update to torch 2.14 (#49199) — 2026-09-30
commit 0.00 QA: fix noisy comment checker (#49184) — 2026-09-30
commit 0.00 Fix command syntax for optimum-cli export (#46451) — 2026-09-29
commit 0.00 Fix Trainer checkpoint resume crashing on CPU with multiple — 2026-09-29
commit 0.00 Fix backslash handling in generate flag values in transforme — 2026-09-29
commit 0.00 Document image_like_kwargs (#49180) — 2026-09-29
commit 0.00 Fix lost call (#49163) — 2026-09-29
commit 0.00 [CB] Fix failing tests discovered when using the B200 (#4917 — 2026-09-29
commit 0.00 [PerceptionLM] Restore test_inputs_embeds overrides to fix f — 2026-09-28
commit 0.00 [MPS] Remove cu_seqlens_k clone workaround for metal-flash-s — 2026-09-28
commit 0.00 No more -hf repo names for ESMC (#49158) — 2026-09-28
commit 0.00 [generation] Encode multimodal data only once (#45783) — 2026-09-28
commit 0.00 [CI] ssh-runner: add optional cache_type input to switch bet — 2026-09-28
commit 0.00 Deprecate the use_mamba_kernels config flag that no longer h — 2026-09-28
commit 0.00 Pick the default flash implementation based on the current h — 2026-09-28
pr 0.00 Smthn smthn The intentionally terse informal wording 2026-10-06
pr 0.00 fix(ci): harden GitHub Actions workflows (#49364) — 2026-10-06
pr 0.00 fix(ci): harden GitHub Actions workflows (#49364) — 2026-10-06
pr 0.00 fix(ci): harden GitHub Actions workflows (#49364) — 2026-10-06
pr 0.00 Revert "[CI] Temporarily disable AMD scheduled CI trigger" ( Terse, conventional revert message with 2026-10-06
pr 0.00 logits_to_keep bool mask for all ForXXX classes, expose last Only a technical title and issue referen 2026-09-15
pr 0.00 fix(ci): harden GitHub Actions workflows (#49357) — 2026-10-06
pr 0.00 fix(ci): harden GitHub Actions workflows (#49357) — 2026-10-06
pr 0.00 Add Molmo2 Mostly the inherited template, with no s 2026-01-23
pr 0.00 ALM testx fIxing... Terse, typo-heavy title and minimal tech 2026-10-06
pr 0.00 CI: Move GPU images to python 3.11 Terse, informal change description with 2026-10-06
pr 0.00 [`distributed`] remaining ep_plan to update Short informal status note, clearly not 2026-09-28
pr 0.00 [`distributed`]: adapt Trainer to work Expert Parallel Only the inherited template and human re 2026-09-16
pr 0.00 fix(ci): harden GitHub Actions workflows (#49342) — 2026-10-06
pr 0.00 fix(ci): harden GitHub Actions workflows (#49342) — 2026-10-06
pr 0.00 fix(ci): harden GitHub Actions workflows (#49342) — 2026-10-06
pr 0.00 [serge] Fix 2 integration tests for model `deepseek_vl_hybri — 2026-10-06
pr 0.00 [serge] Fix 2 integration tests for model `ministral3` faili — 2026-10-06