| commit |
1.00 |
Align image feature dtype in kosmos2 and kosmos2_5 embedding |
|
Commit message contains explicit AI assi |
2026-08-03 |
| commit |
1.00 |
Fix spelling/grammar in model files (batch 1/3) (#47682) |
|
Commit message contains explicit AI assi |
2026-07-31 |
| commit |
1.00 |
Fix spelling/grammar in English docs (q → z) (#47681) |
|
Commit message contains explicit AI assi |
2026-07-31 |
| commit |
1.00 |
Fix spelling/grammar in English docs (h → p) (#47680) |
|
Commit message contains explicit AI assi |
2026-07-31 |
| commit |
1.00 |
Fix spelling/grammar in English docs (a → g) (#47679) |
|
Commit message contains explicit AI assi |
2026-07-31 |
| commit |
1.00 |
Exportable kimi (#47096) |
|
Commit message contains explicit AI assi |
2026-07-31 |
| commit |
1.00 |
[Chat Parsing] Type inline tool-call arguments from the call |
|
Commit message contains explicit AI assi |
2026-07-30 |
| commit |
1.00 |
Add Granite-swa and Granitemoe-swa model support (#47179) |
|
Commit message contains explicit AI assi |
2026-07-29 |
| commit |
1.00 |
byebye torch 2.4 (#47609) |
|
Commit message contains explicit AI assi |
2026-07-29 |
| commit |
1.00 |
Fix CUDA Graph breaking host to device copy from scalar tens |
|
Commit message contains explicit AI assi |
2026-07-29 |
| commit |
1.00 |
CI: use a single function for GH calls (#47474) |
|
Commit message contains explicit AI assi |
2026-07-28 |
| commit |
1.00 |
Kernels and loaders robustification (#47334) |
|
Commit message contains explicit AI assi |
2026-07-28 |
| pr |
1.00 |
Add support for Voxtral-4B-TTS-2603 to transformers |
|
PR body explicitly mentions AI collabora |
2026-04-13 |
| pr |
1.00 |
Support quantization in Dynamo, ONNX and ET exporters |
|
PR body explicitly mentions AI collabora |
2026-08-03 |
| pr |
1.00 |
Add shared ImageProcessingTester |
|
PR body explicitly mentions AI collabora |
2026-08-03 |
| pr |
1.00 |
Follow the hidden states dtype in MaskFormerSwin's shifted-w |
|
Explicitly discloses code agent/AI assis |
2026-08-03 |
| pr |
1.00 |
Do not modify caller-provided tensors in place during forwar |
|
Explicit AI agent assistance disclosure |
2026-08-03 |
| pr |
1.00 |
fix(tokenization): decode byte-fallback tokens in SentencePi |
|
Explicitly states 'Implemented with Clau |
2026-08-03 |
| pr |
1.00 |
Remove stale and redundant _no_split_modules entries |
|
PR body explicitly mentions AI collabora |
2026-07-30 |
| pr |
1.00 |
Fix: Remove redundant @can_return_tuple conflicting with @ca |
|
PR body explicitly mentions AI collabora |
2026-08-03 |
| pr |
1.00 |
Fix feature dtype mismatch in masked_scatter for seven multi |
|
Disclosure line states code agent/AI ass |
2026-07-31 |
| pr |
1.00 |
Import utils compilation fixes |
|
PR body explicitly mentions AI collabora |
2026-08-03 |
| pr |
1.00 |
Executorch exporter fixes |
|
PR body explicitly mentions AI collabora |
2026-07-10 |
| pr |
1.00 |
Apertus 1.5: add model |
|
PR body explicitly mentions AI collabora |
2026-07-31 |
| pr |
1.00 |
[Improvement] Make gated delta rule more explicit and suppor |
|
PR body explicitly mentions AI collabora |
2026-07-29 |
| pr |
1.00 |
Fix infeasible cost matrix errors in hugarian matcher losses |
|
PR body explicitly mentions AI collabora |
2026-08-03 |
| pr |
1.00 |
fix(llama): validate divisibility only when deriving head_di |
|
PR body explicitly mentions AI collabora |
2026-08-02 |
| pr |
1.00 |
fix processor config nested key fallback |
|
PR body explicitly mentions AI collabora |
2026-07-29 |
| pr |
1.00 |
Fix failing tests for granitemoe_swa |
|
PR body explicitly mentions AI collabora |
2026-08-03 |
| pr |
1.00 |
docs(serve): note OpenAI client base_url for multi-model gat |
|
Has 'Co-Authored-By: Claude Fable 5', an |
2026-08-03 |
| pr |
1.00 |
[docs] Fix BatchEncoding documentation inconsistencies |
|
PR body explicitly mentions AI collabora |
2026-07-30 |
| pr |
1.00 |
Add Unlimited OCR |
|
PR body explicitly mentions AI collabora |
2026-06-23 |
| pr |
1.00 |
Remove unused DPT fusion residual parameters |
|
PR body explicitly mentions AI collabora |
2026-08-01 |
| pr |
1.00 |
FlashAttention decode packed-sequence sync fix |
|
PR body explicitly mentions AI collabora |
2026-07-30 |
| pr |
1.00 |
Fix default special token IDs for SigLIP and SigLIP2 configs |
|
PR body explicitly mentions AI collabora |
2026-07-31 |
| pr |
1.00 |
added the chronous 2 first revision code |
|
PR body explicitly mentions AI collabora |
2026-07-31 |
| pr |
1.00 |
[Mistral] Add native tekken tokenizer support to AutoTokeniz |
|
PR body explicitly mentions AI collabora |
2026-07-23 |
| pr |
0.80 |
Update daily CI Docker image to torch 2.13.0 / CUDA 13.0 |
|
Clear AI signature: '🤖 Generated with Cl |
2026-08-03 |
| commit |
0.15 |
Add support for batched Qwen2.5/3-Omni audio generation (#47 |
|
Domain terminology, direct writing, sign |
2026-08-03 |
| pr |
0.15 |
model: Add NVIDIA Canary-1B-v2 to Transformers |
|
Uses domain jargon, compact lists, and r |
2026-06-22 |
| commit |
0.10 |
Remove stale and redundant _no_split_modules entries (#47645 |
|
Free-text is technical, domain-specific, |
2026-08-03 |
| commit |
0.10 |
Update daily CI Docker image to torch 2.13.0 / CUDA 13.0 (#4 |
|
Terse message with technical detail; hum |
2026-08-03 |
| commit |
0.10 |
Fix ImportError in transformers.exporters on torch < 2.8 (#4 |
|
Direct, technical explanation with inlin |
2026-08-03 |
| commit |
0.10 |
fix: correct text input validation logic in 8 multimodal pro |
|
Detailed technical fix, domain terms; na |
2026-07-31 |
| commit |
0.10 |
Improve Trainer DataLoader Controls for Streaming and Multip |
|
Technical, focused, brief PR description |
2026-07-30 |
| commit |
0.10 |
Add AXK2 from SKT (#47528) |
|
Commit uses technical terms, informal to |
2026-07-24 |
| commit |
0.10 |
Fix failing tests for mimo_v2_flash (#47284) |
|
Informal and technical; lists specific t |
2026-07-24 |
| pr |
0.10 |
Add pose estimation keypoint preprocessing to Sapiens2ImageP |
|
Detailed, domain-specific; summary direc |
2026-07-09 |
| pr |
0.10 |
Add qwen3 tts |
|
Domain-specific, concise, uses model and |
2026-03-07 |
| pr |
0.10 |
🚨[wav2vec2] Support attn_implementation=sdpa dispatch |
|
Brief, technical, informal phrasing; rev |
2026-05-25 |
| pr |
0.10 |
Port ESMC and ESMFold2 to Transformers |
|
Informal, personal tone and inline react |
2026-06-04 |
| pr |
0.10 |
Align image feature dtype in kosmos2 and kosmos2_5 embedding |
|
Highly technical, lots of domain detail, |
2026-07-31 |
| pr |
0.10 |
Use explicit nn.Buffer everywhere for modular |
|
'As per the title' intro, but technical |
2026-08-03 |
| pr |
0.10 |
Simplify reverse weight conversion |
|
Informal, terse, has real confusion and |
2026-08-03 |
| pr |
0.10 |
[generate] Stop setting the static cache as an attribute to |
|
Informal, specific discussion, real huma |
2026-08-03 |
| pr |
0.10 |
Minicpm4 |
|
Well-structured engineering PR, domain-s |
2026-08-03 |
| pr |
0.10 |
🚨 [cache] Cropping can only be done with negative values |
|
Informal, long, idiosyncratic, with emoj |
2026-08-03 |
| pr |
0.10 |
Fix DBRX MoE hidden size and expert GLU transposes |
|
Terse, technical, informal language typi |
2026-07-31 |
| pr |
0.10 |
scripts/check_tokenizers: v5 compatibility fixes |
|
Slightly more detailed, but still techni |
2026-08-02 |
| pr |
0.10 |
add deepspeed grad ckpt |
|
Informal tone, domain-specific reference |
2024-04-13 |
| commit |
0.05 |
Executorch exporter fixes (#47243) |
|
Informal, with edits and temporary comme |
2026-08-03 |
| commit |
0.05 |
feat[vLLM]: Support text replacement offsets in the remainin |
|
Domain-specific terms, minimal AI-like p |
2026-08-03 |
| commit |
0.05 |
CI: Add serge review relay workflow and review rules (#47610 |
|
Detailed commit message with domain know |
2026-07-29 |
| pr |
0.05 |
🌐 [i18n-ZH] Translate accelerator_selection.md to Chinese |
|
Straightforward translation PR; notes sh |
2026-08-03 |
| pr |
0.05 |
[docs] Fix failing doctests |
|
Extremely terse, simple description, mat |
2026-07-31 |
| pr |
0.05 |
Add new args in auto-docstring |
|
Extremely brief free-text, classic human |
2026-08-03 |
| pr |
0.05 |
Fix failing tests for axk1 and axk2 |
|
Short, informal and requests review; typ |
2026-08-03 |
| pr |
0.05 |
Fix failing tests for granite_swa and granitemoe_swa |
|
Short, informal and requests review; hum |
2026-08-03 |
| pr |
0.05 |
skip invalid test cases for inkling tests |
|
Informal, uses 'pls' and very brief; hum |
2026-07-23 |
| pr |
0.05 |
Remove `is_custom_code` gate from per-param `_is_hf_initiali |
|
Highly detailed, technical, lengthy with |
2026-07-15 |
| pr |
0.05 |
TP dtensor handle natively the to_local() for deepgem + fp8 |
|
Terse, informal note with dev planning t |
2026-07-30 |
| commit |
0.00 |
Acc fix in xpu (#47500) |
|
Short messages with sign-offs and minor |
2026-08-03 |
| commit |
0.00 |
update Dockerfile for xpu torch2.13 (#47502) |
|
Signed-off commits, concise and practica |
2026-08-03 |
| commit |
0.00 |
Fix failing tests for axk1 and axk2 (#47727) |
|
Technical bugfixes, informal language, d |
2026-08-03 |
| commit |
0.00 |
Fix failing tests for granite_swa and granitemoe_swa (#47723 |
|
Technical fixes, specific test reference |
2026-08-03 |
| commit |
0.00 |
skip invalid test cases for inkling tests (#47493) |
|
Terse commit message; signature shows hu |
2026-08-03 |
| commit |
0.00 |
[docs] Fix BatchEncoding documentation inconsistencies (#476 |
|
Commit message is terse, specific, and c |
2026-08-01 |
| commit |
0.00 |
fix npu check (#47587) |
|
Very terse, code-focused changes with no |
2026-07-31 |
| commit |
0.00 |
Fix feature dtype mismatch in masked_scatter for seven multi |
|
No free-form text or stylistic signals, |
2026-07-31 |
| commit |
0.00 |
[docs] storing and loading chat templates (#47650) |
|
One-word commit messages, no sign of AI |
2026-07-31 |
| commit |
0.00 |
adding amd quark config class changes (#47322) |
|
Brief, repetitive human style; includes |
2026-07-31 |
| commit |
0.00 |
Fix compressed tensors impl (#47652) |
|
Minimal language, purely descriptive com |
2026-07-31 |
| commit |
0.00 |
Fix CodeLlama tokenizer dropping leading whitespace on decod |
|
Technical commit, specific references; h |
2026-07-31 |
| commit |
0.00 |
Hoist special-token lookups in wav2vec2 decode paths and dro |
|
Highly specific technical terminology, n |
2026-07-31 |
| commit |
0.00 |
silencing elastic warning by import distributed lib inside f |
|
Multiple commits with terse, technical h |
2026-07-31 |
| commit |
0.00 |
Fix missing github_utils.py download in PR CI dashboard work |
|
Workflow and technical explanation; huma |
2026-07-31 |
| commit |
0.00 |
better guarding to handle torch compiled with USE_DISTRIBUTE |
|
Terse, technical commit messages typical |
2026-07-31 |
| commit |
0.00 |
Remove gemma4 warnings (#47664) |
|
Extremely brief commit indicative of hum |
2026-07-31 |
| commit |
0.00 |
extend tests/quantization/gptq/test_gptq.py::GPTQTestCUDA an |
|
Commit messages and signatures are terse |
2026-07-30 |
| commit |
0.00 |
Update maintainer list (#47644) |
|
Short, pragmatic commit messages suggest |
2026-07-30 |
| commit |
0.00 |
🚨 Enable SDPA (and other attention backends) for T5 and prop |
|
Informal, incremental commit messages; n |
2026-07-30 |
| commit |
0.00 |
Allow position_ids_start=2 on DataCollatorWithFlattening for |
|
Succinct technical changelog steps and t |
2026-07-30 |
| commit |
0.00 |
Remove Rotary warning (#47642) |
|
Highly informal, terse commit messages; |
2026-07-30 |
| commit |
0.00 |
CI: Log GitHub API diagnostics to stderr (#47635) |
|
Single, brief informal message; clearly |
2026-07-30 |
| commit |
0.00 |
Drop multimodal inputs natively in prepare_inputs_for_genera |
|
Incremental and informal commit messages |
2026-07-30 |
| commit |
0.00 |
Simplify all Rotary modules (#47598) |
|
Commit log is concise, domain-specific, |
2026-07-30 |
| commit |
0.00 |
[docs] response_template when serving (#47626) |
|
Brief commit message in template format, |
2026-07-29 |
| commit |
0.00 |
[docs] MTP support (#47301) |
|
Short, informal notes and fixes, lacks A |
2026-07-29 |
| commit |
0.00 |
[docs] MPS graph cache (#47304) |
|
Sparse, terse commit style, typical huma |
2026-07-29 |
| commit |
0.00 |
Fix GPT-2 c_proj depth scaling initialization (#47459) |
|
Technical details with regression test, |
2026-07-29 |
| commit |
0.00 |
Fix fp8_linear compilability (#47623) |
|
Typos and terse notes, clearly human-gen |
2026-07-29 |
| commit |
0.00 |
[`Kernels`] Refactor function handling (#46883) |
|
Casual language and human commit phrasin |
2026-07-29 |
| commit |
0.00 |
Compressed tensors fp8 (#47216) |
|
Numerous human-typed updates with signed |
2026-07-29 |
| commit |
0.00 |
Vectorize NoRepeatNGramLogitsProcessor and remove its host s |
|
Concise, technical PR title only; no AI |
2026-07-29 |
| commit |
0.00 |
Fix some processors (#47608) |
|
Informal wording, typo, and discussion i |
2026-07-29 |
| commit |
0.00 |
skip fsdp tests when backend is mps (#47601) |
|
Very brief summary and decision, human c |
2026-07-29 |
| commit |
0.00 |
[docs] Exporters (#47374) |
|
Extremely terse commit steps, clearly hu |
2026-07-29 |
| commit |
0.00 |
Fix model parallel device mismatch in `create_bidirectional_ |
|
Technical, concise, uses domain-specific |
2026-07-28 |
| commit |
0.00 |
Fix A.X-K2 fp8 modules_to_not_convert normalization for the |
|
Detailed technical changelog, clear doma |
2026-07-28 |
| commit |
0.00 |
Remove redundant guarding for distributed (#47570) |
|
Terse, domain jargon, and contains typos |
2026-07-28 |
| commit |
0.00 |
Add FSDP plans to all models (#47165) |
|
Terse, technical edits, includes typos/a |
2026-07-28 |
| commit |
0.00 |
Fix modular for mamba packages (#47494) |
|
Very brief, informal commit messages lac |
2026-07-28 |
| commit |
0.00 |
Remove deprecated conversion in Kimi (#47581) |
|
Very terse, informal commit message; cle |
2026-07-27 |
| commit |
0.00 |
CI: use transformers-ci daily workflow with OTEL (#47360) |
|
Terse, informal style with domain-specif |
2026-07-27 |
| commit |
0.00 |
Fix slow tensor path in _check_special_mm_tokens (#47580) |
|
Terse, technical; no AI hallmarks presen |
2026-07-27 |
| commit |
0.00 |
CI: let's run integration failure cron at 10pm (#47537) |
|
Casual language and repetition; human st |
2026-07-27 |
| commit |
0.00 |
Better and more extensive tests for RoPE (#46912) |
|
Informal, fragmented lines; typical huma |
2026-07-27 |
| commit |
0.00 |
Fix mamba2 family decode and simplify all reshape ops (#4756 |
|
Minimal, vague wording; likely human. |
2026-07-27 |
| commit |
0.00 |
Speed up image preprocessing for vision-language models (#47 |
|
Contains domain-specific detail, no AI p |
2026-07-27 |
| commit |
0.00 |
[DiffusionGemma] Cast the decoder padding mask to bool (#472 |
|
Terse, technical; fixes described in a h |
2026-07-27 |
| commit |
0.00 |
Fix NemotronH: Register `"mlp"` in the cache layer-type mapp |
|
Terse, technical, human co-author traile |
2026-07-27 |
| commit |
0.00 |
Fix `config.per_layer_config[layer_type]` lookup speed (#475 |
|
Short, technical; signed off by a human. |
2026-07-27 |
| commit |
0.00 |
General maintenance (#47517) |
|
Brief, informal, lacks AI hallmarks; cle |
2026-07-27 |
| commit |
0.00 |
:rotating_light: Processors update the rest (#46556) |
|
Very informal language, typos, and perso |
2026-07-27 |
| commit |
0.00 |
fixed the benchmark script with DistributedConfig (#47568) |
|
Direct, brief phrasing; contains a typo, |
2026-07-27 |
| commit |
0.00 |
Fix TP inference for tied embedding (#47503) |
|
Terse commit messages, technical jargon, |
2026-07-27 |
| commit |
0.00 |
Fix `causal_conv1d_fn` positional `activation` colliding wit |
|
Short, technical fix message; no AI sign |
2026-07-24 |
| commit |
0.00 |
Refactor all linear attention models to latest best standard |
|
Fragmented, terse commit messages with t |
2026-07-24 |
| commit |
0.00 |
Make tokenization_mistral_common importable without mistral_ |
|
Very short, purely technical; no AI indi |
2026-07-24 |
| commit |
0.00 |
Use --flake-runs=1 in check_bad_commit.py for PR comment CI |
|
Terse, technical; email in trailer, but |
2026-07-24 |
| commit |
0.00 |
Deprecate the old response_schema (#47320) |
|
Direct, technical, no boilerplate or AI- |
2026-07-24 |
| commit |
0.00 |
fix: add pickle support to _LazyConfigMapping for spawn mult |
|
Contains personal signature, technical p |
2026-07-24 |
| commit |
0.00 |
Fix potential ReDoS by escaping tokenizer filename used as r |
|
Technical language, specific context, no |
2026-07-24 |
| commit |
0.00 |
Fix vision position-embedding init width fallback in Phi4Mul |
|
Very brief, technical, no signals of AI- |
2026-07-24 |
| commit |
0.00 |
Delete old deprecations (#47518) |
|
Commit messages are short, informal, and |
2026-07-24 |
| commit |
0.00 |
Run only @slow tests in PR comment CI (#47521) |
|
Terse commit messages, clear human domai |
2026-07-24 |
| commit |
0.00 |
Fix incorrect type hint (#47519) |
|
Simple, human-authored commit message an |
2026-07-24 |
| commit |
0.00 |
Deprecate CB config in gen configuration (#47291) |
|
Quick, shorthand language and typos indi |
2026-07-24 |
| commit |
0.00 |
Allow metal-flash-sdpa for OpenAIPrivacyFilter on MPS (#4674 |
|
Casual, abbreviated style is clearly hum |
2026-07-24 |
| commit |
0.00 |
[Offloading] [Bugfix] Fix fully offloaded model saving (#473 |
|
Concise, domain-specific commit messages |
2026-07-23 |
| commit |
0.00 |
CI: fix torchaudio pinning +proper break in rnnt (#47422) |
|
Informal, terse commit messages with tec |
2026-07-23 |
| commit |
0.00 |
add_axk1 (#46867) |
|
Technical, domain-focused, with typical |
2026-07-23 |
| commit |
0.00 |
Fix Qwen2.5-Omni Token2Wav DiT rotary embedding layout (inte |
|
Detailed, technical explanation with jar |
2026-07-23 |
| commit |
0.00 |
CI: add reproduce mode to serge verify caller (#47492) |
|
Technical writing, straightforward expla |
2026-07-23 |
| commit |
0.00 |
feat[vLLM x v5]: Make audio optional and support multiple au |
|
Typical commit subject/body, no AI phras |
2026-07-23 |
| commit |
0.00 |
[fix][whisper]: fix max_new_tokens handling (#46795) |
|
Brief, domain-specific fix; no sign of A |
2026-07-23 |
| commit |
0.00 |
Isolate MLA KV expansion to make it easier to bypass (#47460 |
|
Technical commit history with human styl |
2026-07-23 |
| commit |
0.00 |
Update serve chat parsing (#46267) |
|
Terse, domain-internal phrasing; no form |
2026-07-23 |
| pr |
0.00 |
[serge] Fix 2 integration tests for model `clvp` failing wit |
|
— |
2026-08-03 |
| pr |
0.00 |
[`Kernels`] Native fallback |
|
Concise, domain-specific, informal notes |
2026-07-29 |
| pr |
0.00 |
[serge] Fix 4 integration tests for model `pvt` failing with |
|
— |
2026-08-03 |
| pr |
0.00 |
Add nvidia/LocateAnything-3B |
|
Demonstrative, detailed, with domain-spe |
2026-07-01 |
| pr |
0.00 |
Re-enable VITS model forward test |
|
Technical summary with concise justifica |
2026-07-06 |
| pr |
0.00 |
Fix RT-DETR v2 converter for current model keys |
|
Detailed technical update with verificat |
2026-07-07 |
| pr |
0.00 |
[Model] Add LingBot-Vision Model |
|
Structured but free-text written in deve |
2026-07-16 |
| pr |
0.00 |
Add support for batched Qwen2.5/3-Omni audio generation |
|
Detailed, technical with rationale and e |
2026-07-09 |
| pr |
0.00 |
clean up reverse_op fixme in compressed_tensors |
|
Casual tone, specific technical focus, a |
2026-08-01 |
| pr |
0.00 |
[docs] Fix links and remove TokenizerFast |
|
Bullet points, minimal explanations, and |
2026-08-03 |
| pr |
0.00 |
Add Fun-ASR-Nano model |
|
Thorough technical explanation, use of m |
2026-05-24 |
| pr |
0.00 |
Acc fix in xpu |
|
Very terse, domain-specific, informal to |
2026-07-23 |
| pr |
0.00 |
Modularize qwen-format vision processors |
|
Free text uses informal tone, some minor |
2026-07-27 |
| pr |
0.00 |
update mistral common version for PR 47507 |
|
Terse, to the point, uses lower-case and |
2026-07-31 |
| pr |
0.00 |
Fix Idefics3/SmolVLM get_number_of_image_patches returning 0 |
|
Detailed, technical, domain-specific, so |
2026-08-03 |
| pr |
0.00 |
[xLSTM] Fix train mode crashes with return_last_states=False |
|
Highly technical, uses informal tone and |
2026-07-03 |
| pr |
0.00 |
DIstributed branch base |
|
Very technical, full of domain jargon, e |
2026-05-28 |
| pr |
0.00 |
[generation] Encode multimodal data only once |
|
Technically detailed, a few grammar laps |
2026-05-05 |
| pr |
0.00 |
Trainer: fix optimizer flat mesh for MoE for native Distribu |
|
Domain-specific jargon, bullet-point exp |
2026-08-03 |
| pr |
0.00 |
update Dockerfile for xpu torch2.13 |
|
Minimal description, uses terse review c |
2026-07-23 |
| pr |
0.00 |
clip_grad_norm avoid O(global_params) replication |
|
Detailed technical analysis, human error |
2026-05-20 |
| pr |
0.00 |
feat[vLLM]: Support text replacement offsets in the remainin |
|
Uses domain-specific notation and a conc |
2026-07-28 |
| pr |
0.00 |
Fix ImportError in transformers.exporters on torch < 2.8 |
|
Direct and technical PR title; no signs |
2026-08-02 |
| pr |
0.00 |
raise error when tp_plan is used instead of distributed_conf |
|
No free-text content provided. |
2026-08-03 |
| pr |
0.00 |
bench: dtensor vs legacy TP |
|
Informal, lowercased 'this PR shows...'; |
2026-08-03 |
| pr |
0.00 |
TP dtensor inference |
|
No free-text content provided. |
2026-07-27 |
| pr |
0.00 |
[serge] Fix 4 integration tests for model `seed_oss` failing |
|
— |
2026-08-02 |
| pr |
0.00 |
[serge] Fix 2 integration tests for model `eomt_dinov3` fail |
|
— |
2026-08-02 |
| pr |
0.00 |
squad_metrics: use precompiled regex for article removal |
|
Only a descriptive, technical PR title w |
2026-08-02 |
| pr |
0.00 |
Fix generate() crash when reusing image_hidden_states with p |
|
PR title uses precise, technical languag |
2026-08-02 |
| pr |
0.00 |
config: raise on missing local config file instead of silent |
|
Technical, terse language; no AI-style p |
2026-08-02 |
| pr |
0.00 |
Fix sliding window cache index off-by-one on wraparound |
|
Technical description with clear, direct |
2026-08-02 |
| pr |
0.00 |
Fix D2H sync in Qwen VL vision encoder |
|
Only the PR title is present; template c |
2026-08-01 |
| pr |
0.00 |
Fix text-generation pipeline not forwarding tokenizer to gen |
|
Template plus minimal, specific title; n |
2026-08-01 |
| pr |
0.00 |
fix npu check |
|
Title-only; contains abbreviation, infor |
2026-07-27 |
| pr |
0.00 |
Build Llama pipeline from the reference SentencePiece normal |
|
Short technical title only; no AI-like f |
2026-08-01 |
| pr |
0.00 |
[serge] Fix 2 integration tests for model `dab_detr` failing |
|
— |
2026-08-01 |
| pr |
0.00 |
[serge] Fix 2 integration tests for model `phi3` failing wit |
|
— |
2026-08-01 |
| pr |
0.00 |
[serge] Fix 2 integration tests for model `swiftformer` fail |
|
— |
2026-07-30 |
| pr |
0.00 |
🚨 [DiffusionGemma] Return router logits and load balancing l |
|
Short, specific title with emoji, no AI |
2026-06-14 |
| pr |
0.00 |
Forward loss kwargs to self.loss_function so gradient accumu |
|
Includes detailed technical explanation, |
2026-07-31 |
| pr |
0.00 |
Add heterogeneous model support (per-layer config and modeli |
|
PR title only, no signs of AI-style phra |
2026-04-09 |
| pr |
0.00 |
Fix Grounding DINO text position ids being truncated to int6 |
|
Concise, technical PR title; template no |
2026-07-31 |
| pr |
0.00 |
Fix SDPA logit corruption for left-padded generation in mult |
|
Free-text is just a title, no AI hallmar |
2026-07-31 |
| pr |
0.00 |
Fix spelling/grammar in model files (batch 1/3) |
|
Title-only; no AI hallmarks or excessive |
2026-07-31 |
| pr |
0.00 |
Fix spelling/grammar in English docs (q → z) |
|
Title-only; clear, direct, and typical e |
2026-07-31 |
| pr |
0.00 |
Fix spelling/grammar in English docs (h → p) |
|
Title-only; concise human summary, no AI |
2026-07-31 |
| pr |
0.00 |
Fix spelling/grammar in English docs (a → g) |
|
Title-only; succinct and natural for a P |
2026-07-31 |
| pr |
0.00 |
Trainer: add predict with generate |
|
Description is technical and concise, fo |
2024-07-31 |
| pr |
0.00 |
Fix spelling/grammar in model files (batch 3/3) |
|
Title-only; no signs of AI generation. |
2026-07-31 |
| pr |
0.00 |
Fix spelling/grammar in model files (batch 2/3) |
|
Title-only; no AI stylistic traits. |
2026-07-31 |
| pr |
0.00 |
Fix spelling/grammar in core library, examples, and utils |
|
Title-only; straightforward and direct, |
2026-07-31 |
