← back to report

huggingface/transformers

// all events — 2026-10-03

Type AI Score Description Actor Reason Date
commit 1.00 [CI] check_bad_commit: use EFS cache to avoid Xet FUSE OOM (
ydshieh
Commit message contains explicit AI assi 2026-10-02
commit 1.00 [ESM/Persimmon/NllbMoe/ESMFold] use safetensors repos to fix
ydshieh
Commit message contains explicit AI assi 2026-10-01
commit 1.00 QA: restore masking_utils export comment with a noqa (#49209
tarekziade
Commit message contains explicit AI assi 2026-09-30
commit 1.00 Replace datasets that no longer load with maintained uploads
davanstrien
Commit message contains explicit AI assi 2026-09-30
commit 1.00 Finish removing the MPS autocast workaround (#49157)
rubenG1009
Commit message contains explicit AI assi 2026-09-29
commit 1.00 Document running the example scripts on Hugging Face Jobs (#
davanstrien
Commit message contains explicit AI assi 2026-09-29
commit 1.00 [CacheHardIntegrationTest] use dedicated safetensors repo to
ydshieh
Commit message contains explicit AI assi 2026-09-29
commit 1.00 [imagegpt/vilt/trocr] fix cache fixture tests: use hf_hub_do
ydshieh
Commit message contains explicit AI assi 2026-09-29
commit 1.00 Fix get_json_schema dropping items/enum for unions of list/d
JoeyTan21
Commit message contains explicit AI assi 2026-09-28
commit 1.00 [Gemma] Fix test_model_7b_fp16_static_cache expected value f
ydshieh
Commit message contains explicit AI assi 2026-09-28
commit 1.00 Restore the Unicode whitespace set in the GPT-SW3 tokenizer
David-Wu1119
Commit message contains explicit AI assi 2026-09-28
commit 1.00 [gemma3n] fix audio test fixture: use hf_hub_download instea
ydshieh
Commit message contains explicit AI assi 2026-09-27
commit 1.00 Remap the legacy Gemma 1 hidden_act in the config post-init
PCfVW
Commit message contains explicit AI assi 2026-09-25
commit 1.00 [NemotronH-Omni] Fix device mismatch in test tensor creation
ydshieh
Commit message contains explicit AI assi 2026-09-25
commit 1.00 Use namespaced dataset ids in docs and PyTorch examples (#49
davanstrien
Commit message contains explicit AI assi 2026-09-25
commit 1.00 [Zamba] Fix associative scan breaking ONNX export and OOM in
ydshieh
Commit message contains explicit AI assi 2026-09-25
commit 1.00 Bump huggingface_hub upper bound to <3.0 (transformers) (#49
Wauplin
Commit message contains explicit AI assi 2026-09-24
commit 1.00 Keep already-decoded array arguments unchanged (#49062)
yonigozlan
Commit message contains explicit AI assi 2026-09-24
commit 1.00 Fix odd head_dim validation for RoPE configurations (#48524)
somuai
Commit message contains explicit AI assi 2026-09-24
pr 1.00 Fix multi-image feature ordering during generation batch exp
DimensionSTP
PR body explicitly mentions AI collabora 2026-10-02
pr 1.00 Preserve masked EOS scores in exponential decay length penal
gitedmond
Explicitly states that AI assistance was 2026-10-02
pr 1.00 Fix encoder repetition penalty source rows after batch expan
gitedmond
Explicitly states that AI assistance was 2026-10-02
pr 1.00 docs: document DynamicCache key_cache/value_cache/legacy_cac
medhansh2808
PR body explicitly mentions AI collabora 2026-09-02
pr 1.00 Composite transparent images in shared RGB conversion and lo
Nikhi00718
Explicitly states that AI assistance was 2026-09-24
pr 1.00 Link the padding side section from SSM and hybrid model docs
jayzuccarelli
PR body explicitly mentions AI collabora 2026-10-02
pr 1.00 [CI] Investigate exit 137 on large model tests (Xet FUSE OOM
ydshieh
Explicitly includes an AI co-author trai 2026-10-01
pr 1.00 Generation cleanup in pipelines
Rocketknight1
Explicitly states that the PR is mostly 2026-10-02
pr 1.00 Composite lm head tp plan
IlyasMoutawwakil
PR body explicitly mentions AI collabora 2026-09-24
pr 1.00 Fix test_all_params_have_gradient for intentionally frozen p
thillai-c
PR body explicitly mentions AI collabora 2026-10-02
pr 1.00 Add TDT loss kernel
ebezzam
Explicitly identifies Claude Code as hav 2026-05-19
pr 1.00 [New model] Add K2 Horizon
boyangaaaaa
PR body explicitly mentions AI collabora 2026-09-27
pr 1.00 Finegrained kernels: unified quantization kernels api (separ
IlyasMoutawwakil
PR body explicitly mentions AI collabora 2026-09-22
pr 1.00 Add Intel XPU CI
regisss
PR body explicitly mentions AI collabora 2026-10-02
pr 1.00 fix precomputed flash kwargs clashing
IlyasMoutawwakil
PR body explicitly mentions AI collabora 2026-08-12
pr 1.00 MoE: shard the experts in every TP/EP plan, and test for it
IlyasMoutawwakil
PR body explicitly mentions AI collabora 2026-09-30
pr 1.00 [generate] Opt-in normalized repetition penalty
captainpete
PR body explicitly mentions AI collabora 2026-09-09
pr 1.00 Upgrade Python Min version to `3.11`
Sai-Suraj-27
PR body explicitly mentions AI collabora 2026-10-02
pr 1.00 Use Siglip2Tokenizer when the Hub tokenizer_class is GemmaTo
jayzuccarelli
PR body explicitly mentions AI collabora 2026-10-02
pr 1.00 [Nemotron3Diarization] Pad the last streaming chunk by n_fft
happyarts
PR body explicitly mentions AI collabora 2026-09-30
pr 1.00 Apertus 1.5: add model
RaphaelKreft
PR body explicitly mentions AI collabora 2026-07-31
pr 0.91 [security] Add SSRF validation to load_image URL fetch
AUTHENSOR
Highly structured security report uses g 2026-07-10
pr 0.30 Add _loss_is_scaled_for_ga to allow custom trainers to contr
abigailllr
Generic polished wording, though concise 2026-01-31
pr 0.24 docs: Add advanced masking example for shared prefixes in Mo
sayan5069
Polished explanatory prose, but it inclu 2026-10-01
pr 0.22 test: Skip symlink test on Windows without privilege (fixes
CJstate
Detailed platform-specific test informat 2026-09-07
pr 0.20 Fix loading SentencePiece tokenizers with empty precompiled_
Pushpak731
Structured technical explanation, but no 2026-09-21
pr 0.12 modular cleanup
itazap
Structured planning text is somewhat pol 2026-10-02
pr 0.12 Fix RoPE corruption from TF32 matmul dispatch on B200 (#4842
adityasharmaaaaa
Polished technical prose, but no clear A 2026-09-01
pr 0.12 Fix: Skip weight initialization for quantized int8 models
imjbassi
Specific PyTorch details and concise wor 2025-07-17
pr 0.12 Fix Trainer step accounting when GradScaler skips an optimiz
gitedmond
Highly polished but contains concrete de 2026-10-01
pr 0.10 Add heterogeneous model support (per-layer config and modeli
eladsegal
Detailed, domain-specific engineering de 2026-04-09
pr 0.10 Preserve full request history for continuous batching logits
Robertboy18
Specific implementation context and casu 2026-09-01
pr 0.08 Warm the page cache before sharded loading (opt-in via `HF_S
qgallouedec
Highly specific performance measurements 2026-08-23
pr 0.08 Log learning rate
muellerzr
Informal, concise description with a con 2023-11-22
pr 0.08 Add Trainer.loss_is_scaled_for_ga to declare whether compute
qgallouedec
Specific Trainer behavior, issue referen 2026-10-01
pr 0.08 [generation] Fix encoder repetition penalty for expanded bat
lindicaphxag-tech
Specific bug mechanics and targeted regr 2026-10-02
pr 0.08 Serialize MPS transfers in async model loading (fix #48029)
Irfanwani
Highly detailed technical report, but la 2026-09-28
pr 0.08 Qualcomm AI Engine Direct - Support QNN quantization and low
shewu-quic
Technical implementation details and pro 2026-09-29
pr 0.08 fix(hf_argparser): support bare list dataclass fields
aniketkrs
Detailed technical diagnosis and verific 2026-09-25
pr 0.08 This PR generalizes continuous batching graph support from C
sywangyi
Detailed accelerator-specific changelog 2026-08-27
pr 0.07 Fix MusicGen decoder dropout defaults (#49094)
Mohit-25-tech
Specific regression results and caveats 2026-10-01
pr 0.07 Fix AutoTokenizer backend selection for CohereLabs tiny-aya
Subodh-17
Clear technical explanation and concrete 2026-09-29
pr 0.06 [CI] check_bad_commit: use EFS cache to avoid Xet FUSE OOM (
ydshieh
Detailed operational context and concret 2026-10-02
pr 0.06 Prefetch only the byte spans each rank will read
qgallouedec
Contains unusually specific benchmarks a 2026-09-07
pr 0.06 Add native LFM2-Audio support
kadirnar
Highly polished technical summary, but c 2026-08-24
pr 0.06 Add qwen3 tts
ShahVandit
Structured model summary is polished but 2026-03-07
pr 0.04 :rotating_light: Infer config class default fields without i
zucchini-nlp
Domain-specific details and informal typ 2026-08-25
pr 0.04 Add WeatherNext 2 weather forecasting model
kashif
Detailed domain-specific explanation wit 2026-08-10
pr 0.04 Don't forward text-level flash-attention kwargs to multimoda
Abdennacer-Badaoui
Specific implementation details and info 2026-10-01
pr 0.04 Support per-layer cache configuration
eladsegal
Concise, repository-specific description 2026-08-21
commit 0.03 Add Trainer.loss_is_scaled_for_ga to declare whether compute
qgallouedec
Technical, repository-specific wording a 2026-10-02
pr 0.03 Enable model test on upcoming `torch_tpu` backend
tengomucho
Technical specifics, grammatical slips, 2026-09-30
pr 0.03 [CI] Temporarily disable AMD scheduled CI trigger
ydshieh
Terse, specific CI change with natural e 2026-10-02
pr 0.03 Add Molmo2
SangbumChoi
Minimal free text with no clear AI-style 2026-01-23
pr 0.03 Support fdma for models with attention bias
LoserCheems
Free text is minimal; remaining overview 2025-10-27
pr 0.03 Fix SAM3 video reconditioning masks
jrile018
Concise, issue-specific description with 2026-09-30
pr 0.03 Fix CJK decoding for LlamaTokenizerFast by loading full pipe
MUKUL-PRASAD-SIGH
Specific tokenizer internals, issue refe 2026-10-02
pr 0.03 Route byte-level llama tokenizers to TokenizersBackend
subin9
Detailed tokenizer behavior, examples, a 2026-07-02
pr 0.03 handle transpose during model loading with TP (Qwen3-VL-MoE)
3outeille
Contains technical specifics, repetition 2026-08-27
pr 0.03 Fix dtype mismatch in in modeling_llava_next
Godkunn
Short, concrete fix description with loc 2025-12-21
commit 0.02 🚨 Select the OWLv2 image query by objectness (#49200)
qgallouedec
Specific technical change with a normal 2026-10-02
commit 0.02 Support per-layer cache configuration (#48178)
eladsegal
Long iterative history contains specific 2026-10-02
commit 0.02 Remove rotaries deprecation cycle (#49228)
Cyrilvallez
Terse, informal commit-style wording sug 2026-10-02
commit 0.02 [CB] Make CB more device agnostic and add support for XPU (#
remi-or
Detailed domain-specific summary and ite 2026-10-01
pr 0.02 Do RoPE with mul instead of matmul
Rocketknight1
Specific technical rationale and natural 2026-10-02
pr 0.02 [docs] Update Spaces
stevhliu
Brief informal wording and grammatical e 2026-10-01
pr 0.02 🚨 Select the OWLv2 image query by objectness
qgallouedec
Specific experiment results, jargon, and 2026-09-30
pr 0.02 Clip grad norm dtensor
michaelbenayoun
Concise, repository-specific description 2026-09-29
pr 0.02 Revert "[CI] ssh-runner: add optional cache_type input to sw
ydshieh
Brief, repository-specific revert explan 2026-10-02
pr 0.02 Add MOSS-Transcribe-Diarize model
itazap
Terse, placeholder-heavy text and inform 2026-08-28
pr 0.02 Rename GradientCheckpointingLayer and supports_gradient_chec
qgallouedec
Specific internal API details and review 2026-10-02
pr 0.02 Rename gradient_checkpointing to activation_checkpointing in
qgallouedec
Detailed compatibility behavior and doma 2026-10-02
pr 0.02 FSDP x TP Saving / Loading
michaelbenayoun
Contains natural grammatical errors and 2026-09-14
pr 0.02 FSDP x TP Saving / Loading, Optimizer
michaelbenayoun
Concise, domain-specific optimizer descr 2026-10-02
pr 0.02 fix nested set_attn_implementation
eustlb
Terse issue-focused description and info 2026-09-27
pr 0.02 Remove deprecation cycle on mask and cache
Cyrilvallez
Terse, informal wording and review notes 2026-10-01
pr 0.02 act offload with full overlap
stas00
Short, informal follow-up description wi 2026-09-01
commit 0.01 Revert "[CI] ssh-runner: add optional cache_type input to sw
ydshieh
Concise revert message with a specific c 2026-10-02
commit 0.01 Don't forward text-level flash-attention kwargs to multimoda
Abdennacer-Badaoui
Terse commit notes and an informal CI re 2026-10-02
commit 0.01 [`GTE`] Add memory mixin (#49234)
vasqu
Terse, informal wording is characteristi 2026-10-01
commit 0.01 [docs] TP for PEFT adapters (#49060)
stevhliu
Brief, informal commit notes lack AI-sty 2026-10-01
commit 0.01 correct final metrics due to resuming from checkpoint (#4920
SunMarc
Concise wording is typical of a human co 2026-10-01
commit 0.01 🚨 Attention 🚨 Deprecate "paged|" prefix for SDPA and flash
remi-or
Terse commit sequence and domain-specifi 2026-10-01
commit 0.01 🚨 Attention 🚨 Make regular attention support CB (#49101)
remi-or
Informal review notes, typos, and co-aut 2026-10-01
commit 0.01 🚨 [CB] 🚨 Fuse update for index and block table path (#49088)
remi-or
Minimal phrase is a typical human squash 2026-10-01
commit 0.01 [Refactor] Make some flash-attention utils more readable (#4
remi-or
Short technical change list and informal 2026-10-01
commit 0.01 🚨 [CB] 🚨 Little fixes before removing "paged" (#49069)
remi-or
Terse technical notes and shorthand are 2026-10-01
commit 0.01 fix the slack API call (#49220)
tarekziade
Simple wording and a specific workflow f 2026-10-01
pr 0.01 Fix minicpm 4.6V video batching and CI
zucchini-nlp
Terse wording, abbreviations, and errors 2026-10-02
pr 0.01 Fix pp chart2table inference
zucchini-nlp
Extremely terse wording and casual typo 2026-10-01
pr 0.01 Allow registering new processing backends
zucchini-nlp
Informal wording, typos, and specific ba 2026-09-21
pr 0.01 Adding features like Tokenizer evaluation/benchmarking
rdharshinir
Brief informal title and response provid 2025-04-25
pr 0.01 Reapply modular examples
Cyrilvallez
Very terse, informal wording is consiste 2026-10-01
pr 0.01 Remove some stuff that does not exist
Cyrilvallez
Very terse and informal wording is chara 2026-10-01
pr 0.01 Remove rotaries deprecation cycle
Cyrilvallez
Very brief and informal description with 2026-10-01
pr 0.01 🚨 Deprecate low use models
molbap
Casual phrasing, typo, and terse model l 2026-09-25
pr 0.01 Return router logits from every MoE model that computes them
qgallouedec
Highly domain-specific details and infor 2026-09-18
pr 0.01 fix: DataCollatorWithFlattening incompatible with Tensor inp
sudhanshu746
Awkward grammar, typo-like formatting, a 2024-10-20
pr 0.01 Update kernelize use_fallback
SunMarc
Brief, specific technical description wi 2026-10-01
pr 0.01 [CB] Make CB more device agnostic and add support for XPU
remi-or
Informal acknowledgements and concrete p 2026-09-28
pr 0.01 Little micro-optimization for FlashAttention
Rocketknight1
Casual phrasing and highly specific perf 2026-09-25
pr 0.01 [docs] TP for PEFT adapters
stevhliu
Brief, natural project-specific wording 2026-09-23
commit 0.00 Fix pp chart2table inference (#49225)
zucchini-nlp
Informal typo-filled notes are character 2026-10-02
commit 0.00 fix precomputed flash kwargs clashing (#47928)
IlyasMoutawwakil
Extremely terse and informal commit mess 2026-10-02
commit 0.00 Reapply modular examples (#49230)
Cyrilvallez
Brief, informal commit notes with no AI- 2026-10-02
commit 0.00 Remove some stuff that does not exist (#49232)
Cyrilvallez
Terse wording and informal phrasing indi 2026-10-02
commit 0.00 Remove deprecation cycle on mask and cache (#49231)
Cyrilvallez
Minimal, informal revert description. 2026-10-02
commit 0.00 Dev update (#49212)
vasqu
— 2026-09-30
commit 0.00 model: Add GTE to Transformers (#48416)
harshaljanjani
— 2026-09-30
commit 0.00 [Nemotron3Diarization] fix streaming last stft frame dropped
eustlb
— 2026-09-30
commit 0.00 Fix missing router_logits in Qwen3.5-MoE and other MoE model
zucchini-nlp
— 2026-09-30
commit 0.00 Fix MiniMax M3 partial 3D vision rotary embeddings (#49164)
cuichenx
— 2026-09-30
commit 0.00 Add Strix Halo (gfx1151) Atlas Inference Hub-kernel path for
AzeezIsh
— 2026-09-30
commit 0.00 Fix MPS GQA version gating (#49210)
Isalia20
— 2026-09-30
commit 0.00 Fix additional_special_tokens data loss with extra_special_t
erichanwang
— 2026-09-30
commit 0.00 Fix stale _added_tokens_encoder entries in cpmant and wav2ve
ishan-1010
— 2026-09-30
commit 0.00 Fix deepstack features for mixed-input (#49177)
zucchini-nlp
— 2026-09-30
commit 0.00 gguf user defined tokens (#49008)
SunMarc
— 2026-09-30
commit 0.00 [`Kernels`] Sync mamba version (#49205)
vasqu
— 2026-09-30
commit 0.00 Map bare list, tuple and dict annotations to the right JSON
825pranav
— 2026-09-30
commit 0.00 Fix SequenceBiasLogitsProcessor edge cases: token id 0 and p
lucaluo925
— 2026-09-30
commit 0.00 Fail fast on eval OOM under `auto_find_batch_size` (#49198)
qgallouedec
— 2026-09-30
commit 0.00 [MPS] Let mps sdpa handle grouped query attention directly (
Isalia20
— 2026-09-30
commit 0.00 [Parakeet] Convert NeMo's stochastic depth to layerdrop (#49
Deep-unlearning
— 2026-09-30
commit 0.00 Video processors - general maintenance (#48251)
zucchini-nlp
— 2026-09-30
commit 0.00 update to torch 2.14 (#49199)
sywangyi
— 2026-09-30
commit 0.00 QA: fix noisy comment checker (#49184)
tarekziade
— 2026-09-30
commit 0.00 Fix command syntax for optimum-cli export (#46451)
Ahwar
— 2026-09-29
commit 0.00 Fix Trainer checkpoint resume crashing on CPU with multiple
neevmodh
— 2026-09-29
commit 0.00 Fix backslash handling in generate flag values in transforme
JHC56
— 2026-09-29
commit 0.00 Document image_like_kwargs (#49180)
guarin
— 2026-09-29
commit 0.00 Fix lost call (#49163)
molbap
— 2026-09-29
commit 0.00 [CB] Fix failing tests discovered when using the B200 (#4917
remi-or
— 2026-09-29
commit 0.00 [PerceptionLM] Restore test_inputs_embeds overrides to fix f
ydshieh
— 2026-09-28
commit 0.00 [MPS] Remove cu_seqlens_k clone workaround for metal-flash-s
Isalia20
— 2026-09-28
commit 0.00 No more -hf repo names for ESMC (#49158)
Rocketknight1
— 2026-09-28
commit 0.00 [generation] Encode multimodal data only once (#45783)
zucchini-nlp
— 2026-09-28
commit 0.00 [CI] ssh-runner: add optional cache_type input to switch bet
ydshieh
— 2026-09-28
commit 0.00 Deprecate the use_mamba_kernels config flag that no longer h
albertvillanova
— 2026-09-28
commit 0.00 Pick the default flash implementation based on the current h
Abdennacer-Badaoui
— 2026-09-28
commit 0.00 Auto generate model inits (#47829)
guarin
— 2026-09-28
commit 0.00 Skip flash tests that fall back to a hub kernel when kernels
Abdennacer-Badaoui
— 2026-09-28
commit 0.00 Fix UMT5 decoder self-attention not being causal (#49135)
the-cross-art
— 2026-09-28
commit 0.00 [AMD] Fix some integration tests (#49153)
Abdennacer-Badaoui
— 2026-09-28
commit 0.00 qwen3 models map to wrong tokenizer class on the hub (#49116
itazap
— 2026-09-28
commit 0.00 Fix exporters import on torch < 2.9 (is_contiguous_or_false)
ydshieh
— 2026-09-26
commit 0.00 [serge] Fix 2 integration tests for model `jamba` failing wi
sergereview[bot]
— 2026-09-25
commit 0.00 Use grouped_mm on TPU devices under torch.compile (#49097)
salkan0
— 2026-09-25
commit 0.00 Keep the eos ids the config declares (#49082)
albertvillanova
— 2026-09-25
commit 0.00 [serge] Fix 2 integration tests for model `pvt_v2` failing w
sergereview[bot]
— 2026-09-25
commit 0.00 [serge] Fix 2 integration tests for model `hy_v3` failing wi
sergereview[bot]
— 2026-09-25
commit 0.00 [serge] Fix 2 integration tests for model `cvt` failing with
sergereview[bot]
— 2026-09-25
commit 0.00 Fix MaskFormerSwin attention mask dtype to follow hidden sta
kaixuanliu
— 2026-09-25
commit 0.00 Fix RGB early-return skipping PNG tRNS compositing (#49005)
cs-fisha
— 2026-09-25
commit 0.00 [Executorch] Add MLX recipe (#48910)
metascroy
— 2026-09-25
commit 0.00 Add image processing tester init (#48829)
guarin
— 2026-09-24
commit 0.00 Add `Trainer.end` (#48875)
qgallouedec
— 2026-09-24
commit 0.00 🚨 [ROCm] gpt-oss: route FA3 to aiter-flash-attn, generate RO
Abdennacer-Badaoui
— 2026-09-24
commit 0.00 Keep `attention_mask` as `None` in OPT's causal mask creatio
jiqing-feng
— 2026-09-24
commit 0.00 Fix NaN in Parakeet eager attention with padded batches (#49
ArthurZucker
— 2026-09-24
commit 0.00 [serge] Fix OOM in Moshi integration tests with MemoryCleanu
sergereview[bot]
— 2026-09-24
commit 0.00 Fix assisted eos token condition (#49075)
Cyrilvallez
— 2026-09-24
commit 0.00 Fix XPU Docker image build (#49073)
regisss
— 2026-09-24
commit 0.00 Add support for Nemotron Omni (#46509)
meatybobby
— 2026-09-24
commit 0.00 Honor `config.output_router_logits` in the MoE VLM wrappers
qgallouedec
— 2026-09-23
commit 0.00 fix(ci): harden GitHub Actions workflows (#49057) (#49059)
hf-security-analysis[bot]
— 2026-09-23
commit 0.00 [Nemotron3Diarization] nit: hub pr merged to main (#49065)
eustlb
— 2026-09-23
pr 0.00 [serge] Fix 3 integration tests for model `deepseek_vl_hybri
sergereview[bot]
— 2026-10-02
pr 0.00 fix(ci): harden GitHub Actions workflows (#49254)
hf-security-analysis[bot]
— 2026-10-02
pr 0.00 fix(ci): harden GitHub Actions workflows (#49254)
hf-security-analysis[bot]
— 2026-10-02
pr 0.00 Accidentally created, closed
jayzuccarelli
Terse, informal wording strongly suggest 2026-10-02
pr 0.00 [serge] Fix 2 integration tests for model `fuyu` failing wit
sergereview[bot]
— 2026-10-02
pr 0.00 [serge] Fix 2 integration tests for model `pvt_v2` failing w
sergereview[bot]
— 2026-10-01
pr 0.00 Why override when you can fix in `Mixin`
zucchini-nlp
— 2026-09-25
pr 0.00 fix(utils): support tuple and set typed fields in HfArgument
Gracy769
— 2026-08-30
pr 0.00 [`GTE`] Add memory mixin
vasqu
Extremely terse and informal wording is 2026-10-01
pr 0.00 Add chat input sanitization
Rocketknight1
— 2026-07-17
pr 0.00 [`distributed`] Add expert-parallel token dispatch, default
3outeille
— 2026-09-16
pr 0.00 [`distributed`] decoupled tp_plan and ep_plan
3outeille
— 2026-09-16
pr 0.00 [`distributed`] Add dense and expert device mesh
3outeille
— 2026-09-16
pr 0.00 [`distributed`]: adapt Trainer to work Expert Parallel
3outeille
— 2026-09-16