ceos/unsloth
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions
1,350 commits 391 branches 58 tags 304 MiB
Commit graph

1,350 commits

This branch
This branch
All branches
Author SHA1 Message Date
Daniel Han
d00ebe5fe6 Update rl_replacements.py 2025-06-21 22:20:32 -07:00
Daniel Han
8aa0a2dc37 Fix DAPO, TRL 0.19.0 2025-06-21 22:14:21 -07:00
simpissa
3e75f21086 Fix for grpo_compute_loss_slow (#2702) 
* slice last logit

* move slicing
 2025-06-21 21:58:06 -07:00
Daniel Han
a9b70d94e4 Mistral Small 3.2 2025-06-21 06:44:14 -07:00
amrothemich
dab00ef9a6 Update pyproject.toml (#2778) 
Switched pyproject license to dictionary type
 2025-06-21 02:44:24 -07:00
Michael Han
09ae01fa91 Merge pull request #2780 from rolandtannous/fix/gemma3-grpo-self-llm 
Fix AttributeError in GRPO trainer for models without llm attribute
 2025-06-20 21:15:54 -07:00
Roland Tannous
17563176d9 Fix Gemma3ForCausalLm does not have attribute self.llm 2025-06-21 01:07:32 +00:00
Roland Tannous
4a70f8e880 Additional tests for unsloth-zoo PR#174 2025-06-21 00:22:00 +00:00
Daniel Han
7e7510daf6 Merge branch 'main' of https://github.com/unslothai/unsloth 2025-06-20 06:30:43 -07:00
Daniel Han
9fbb88479c Update pyproject.toml 2025-06-20 06:30:35 -07:00
marcandrelarochelle
35d37ea567 Fix TRL 1.8.2 (#2774) 
* Fix for TRL 1.8.2

Regex matching LLM initialization

* Update Regex
 2025-06-20 06:28:58 -07:00
Daniel Han
1c7a9f46dc Update __init__.py 2025-06-20 06:13:45 -07:00
Daniel Han
a1e12d4243 Fix bugs 2025-06-20 06:09:03 -07:00
Datta Nimmaturi
83f20be8ad Enable vLLM to share memory space (#2712) 
* vLLM sleep once generation is done

* Make enable_sleep_model configurable

* Make default to false

Signed-off-by: datta0 <venkatadattasainimmaturi@gmail.com>

* Force standby under environment variable

---------

Signed-off-by: datta0 <venkatadattasainimmaturi@gmail.com>
 2025-06-19 04:04:14 -07:00
Edd
3a7cf48532 Fix renaming on other model than Llama (#2762) 2025-06-18 13:38:36 -07:00
leopardracer
8a192b7d72 Fix Typos in Documentation and Comments (#2721) 
* Update ocr_eval.md

* Update backward.py
 2025-06-17 04:34:51 -07:00
pluesclues
0270b7166a Reward modeling update (There seems to be another patch) (#2710) 
* Update llama.py, sequence_classifcaiton update

* Update llama.py, adapting to original commit

* Update llama.py, for seqeuence classifcation update

* Update llama.py, added transformer import

* Update llama.py, dealt with output weight

* Update llama.py, renamed it peft model fast forward

* Update llama.py, set up is classification varaiable

* Update llama.py, updated lora dict to initialize sequence classification object

* Update llama.py, gets model name correctly before Lora dict is initialized

* Update llama.py, Task_type_SEQ_CLS doesnt work but it does work with Task_type.CAUSAL_LM
 2025-06-17 04:33:45 -07:00
Michael Han
701692fcfa Update issue templates 
Adding Reddit link
 2025-06-12 01:23:36 -07:00
Roland Tannous
d0287bc596 tests for additional merge fix unsloth zoo pr 163 (#2719) 
* tests for additional merge fix unsloth zoo pr 163

* fixed load_dataset indent in mistral perplexity test file
 2025-06-11 14:08:41 -07:00
Daniel Han
4573105fc8 Versioning 2025-06-10 06:51:07 -07:00
user799595
a497f8878b Making protobuf version more flexible (#2637) 
* Making protobuf version more flexible

* Update pyproject.toml

* Update pyproject.toml

---------

Co-authored-by: Daniel Han <danielhanchen@gmail.com>
 2025-06-10 04:13:25 -07:00
Daniel Han
737dc347d4 Update pyproject.toml 2025-06-10 04:04:25 -07:00
Lei Zhenyuan
14cab69d43 add support for torch270 (#2709) 2025-06-10 03:59:15 -07:00
Daniel Han
d463db673d versioning 2025-06-06 05:46:49 -07:00
Salpingopharyngeus
5cd191b520 Ignore None to Subprocess_Commands (#2680) 
Ignores none params when building the subprocess_command for vllm. As none values stop vllm from deploying properly, as --quantize will be passed with none if quantization type isn't specified in the model name.
 2025-06-05 01:25:12 -07:00
DoubleMathew
e160305b66 Update prepare 4d causal attention call (#2678) 2025-06-04 12:58:50 -07:00
Daniel Han
64e07fc306 Update rl.py 2025-06-03 00:07:52 -07:00
DoubleMathew
c1076ed540 patch sft_trainer to favor max_seq_length over max_length in config (#2669) 2025-06-03 00:06:44 -07:00
DoubleMathew
c33bb76972 unsloth checkpointing fix for latest transformers==4.52.x (#2674) 2025-06-03 00:06:06 -07:00
Roland Tannous
7677750b67 reroute merge logic language models + comprehensive tests + eval kits (#2673) 2025-06-02 20:32:57 -07:00
RunFMe
d80e8a5cd8 Fix batched generation for prompts of different lengths (#2216) 
* fix ignoring of attention mask after prefill stage in decoding

* update naming to avoid confusion

---------

Co-authored-by: Неизвестный Пользователь722497 <dolegosmirnov@sberbank.ru>
 2025-06-02 03:59:10 -07:00
Michael Han
cb07a3608b Merge pull request #2662 from Datta0/model_param_fix 
Fix quant model param fetch regex
 2025-06-01 04:19:12 -07:00
datta0
e6b1a3703d Make replacement logic conscise 2025-06-01 05:57:43 +00:00
Michael Han
0ff1996273 Update issue templates 2025-05-31 14:38:55 -07:00
datta0
fa98cee8f4 Fix quant model param fetch regex 2025-05-31 18:52:46 +00:00
Daniel Han
c7047d014f DeepSeek R1 Qwen 2025-05-30 01:38:53 -07:00
Daniel Han
65825aa571 Bug fixes (#2651) 
* Update vision.py

* Update vision.py

* Update vision.py

* Update vision.py

* Update vision.py

* Update vision.py

* Update loader.py

* check

* Update _utils.py

* Update loader.py

* Update loader.py

* Remove prints

* Update README.md

typo

* Update _utils.py

* Update _utils.py

* versioning

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update vision.py

* HF Transfer

* fix(utils): add missing importlib import to fix NameError (#2134)

This commit fixes a NameError that occurs when `importlib` is referenced in _utils.py
without being imported, especially when UNSLOTH_USE_MODELSCOPE=1 is enabled.
By adding the missing import statement, the code will no longer throw a NameError.

* Add QLoRA Train and Merge16bit Test (#2130)

* add reference and unsloth lora merging tests

* add test / dataset printing to test scripts

* allow running tests from repo root

* add qlora test readme

* more readme edits

* ruff formatting

* additional readme comments

* forgot to add actual tests

* add apache license

* Update pyproject.toml

* Update vision.py

* Update vision.py

* Update vision.py

* Update vision.py

* Update loader.py

* Update loader.py

* Revert

* Update vision.py

* Update vision.py

* Update vision.py

* Update vision.py

* Update vision.py

* Bug fix

* Update mapper.py

* check SDPA for Mistral 3, Pixtral

* Update vision.py

* Versioning

* Update rl_replacements.py

* Update README.md

* add model registry

* move hf hub utils to unsloth/utils

* refactor global model info dicts to dataclasses

* fix dataclass init

* fix llama registration

* remove deprecated key function

* start registry reog

* add llama vision

* quant types -> Enum

* remap literal quant types to QuantType Enum

* add llama model registration

* fix quant tag mapping

* add qwen2.5 models to registry

* add option to include original model in registry

* handle quant types per model size

* separate registration of base and instruct llama3.2

* add QwenQVQ to registry

* add gemma3 to registry

* add phi

* add deepseek v3

* add deepseek r1 base

* add deepseek r1 zero

* add deepseek distill llama

* add deepseek distill models

* remove redundant code when constructing model names

* add mistral small to registry

* rename model registration methods

* rename deepseek registration methods

* refactor naming for mistral and phi

* add global register models

* refactor model registration tests for new registry apis

* add model search method

* remove deprecated registration api

* add quant type test

* add registry readme

* make llama registration more specific

* clear registry when executing individual model registration file

* more registry readme updates

* Update _auto_install.py

* Llama4

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Synthetic data

* Update mapper.py

* Xet and Synthetic

* Update synthetic.py

* Update loader.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update pyproject.toml

* Delete .gitignore

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update _utils.py

* Update pyproject.toml

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update chat_templates.py

* Seasame force float16 / float32

* Fix Seasame

* Update loader.py

* Update vision.py

* Update vision.py

* Update vision.py

* Update loader.py

* is_multimodal

* Update loader.py

* Update loader.py

* Update loader.py

* Update loader.py

* Update vision.py

* Update vision.py

* Update vision.py

* UNSLOTH_DISABLE_STATIC_GENERATION

* Update vision.py

* Auto vision detection

* Sesame

* Whisper

* Update loader.py

* Update loader.py

* Update loader.py

* Update mapper.py

* Update vision.py

* Update vision.py

* Update vision.py

* Update vision.py

* Update vision.py

* Update vision.py

* Update loader.py

* Update loader.py

* Update loader.py

* Update loader.py

* Update _utils.py

* Update rl.py

* versioning

* Update rl.py

* Update rl.py

* Update rl.py

* Update rl.py

* Update rl.py

* logging

* Update pyproject.toml

* Update rl.py

* versioning

* Update rl.py

* Update rl.py

---------

Co-authored-by: Jack Shi Wei Lun <87535974+jackswl@users.noreply.github.com>
Co-authored-by: naliazheli <nalia0316@gmail.com>
Co-authored-by: jeromeku <jerome.ku@gmail.com>
Co-authored-by: Michael Han <107991372+shimmyshimmer@users.noreply.github.com>
 2025-05-29 09:59:29 -07:00
DoubleMathew
ebfa3df9c6 Fix SFTtraining for new trl (#2647) 
* fix sft training with trl>0.15.2 with trl DataCollator

* Update fix to accomodate both trl and transformers DataCollatorForLanguageModeling
 2025-05-28 11:55:48 -07:00
Daniel Han
5d90c8303f Latest TRL, GRPO + Bug fixes (#2645) 
* Update vision.py

* Update vision.py

* Update vision.py

* Update vision.py

* model_type_arch

* Update vision.py

* Update vision.py

* Update vision.py

* Update vision.py

* Update vision.py

* Update vision.py

* Update loader.py

* check

* Update _utils.py

* Update loader.py

* Update loader.py

* Remove prints

* Update README.md

typo

* Update _utils.py

* Update _utils.py

* versioning

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update vision.py

* HF Transfer

* fix(utils): add missing importlib import to fix NameError (#2134)

This commit fixes a NameError that occurs when `importlib` is referenced in _utils.py
without being imported, especially when UNSLOTH_USE_MODELSCOPE=1 is enabled.
By adding the missing import statement, the code will no longer throw a NameError.

* Add QLoRA Train and Merge16bit Test (#2130)

* add reference and unsloth lora merging tests

* add test / dataset printing to test scripts

* allow running tests from repo root

* add qlora test readme

* more readme edits

* ruff formatting

* additional readme comments

* forgot to add actual tests

* add apache license

* Update pyproject.toml

* Update vision.py

* Update vision.py

* Update vision.py

* Update vision.py

* Update loader.py

* Update loader.py

* Revert

* Update vision.py

* Update vision.py

* Update vision.py

* Update vision.py

* Update vision.py

* Bug fix

* Update mapper.py

* check SDPA for Mistral 3, Pixtral

* Update vision.py

* Versioning

* Update rl_replacements.py

* Update README.md

* add model registry

* move hf hub utils to unsloth/utils

* refactor global model info dicts to dataclasses

* fix dataclass init

* fix llama registration

* remove deprecated key function

* start registry reog

* add llama vision

* quant types -> Enum

* remap literal quant types to QuantType Enum

* add llama model registration

* fix quant tag mapping

* add qwen2.5 models to registry

* add option to include original model in registry

* handle quant types per model size

* separate registration of base and instruct llama3.2

* add QwenQVQ to registry

* add gemma3 to registry

* add phi

* add deepseek v3

* add deepseek r1 base

* add deepseek r1 zero

* add deepseek distill llama

* add deepseek distill models

* remove redundant code when constructing model names

* add mistral small to registry

* rename model registration methods

* rename deepseek registration methods

* refactor naming for mistral and phi

* add global register models

* refactor model registration tests for new registry apis

* add model search method

* remove deprecated registration api

* add quant type test

* add registry readme

* make llama registration more specific

* clear registry when executing individual model registration file

* more registry readme updates

* Update _auto_install.py

* Llama4

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Synthetic data

* Update mapper.py

* Xet and Synthetic

* Update synthetic.py

* Update loader.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update pyproject.toml

* Delete .gitignore

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update _utils.py

* Update pyproject.toml

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update chat_templates.py

* Seasame force float16 / float32

* Fix Seasame

* Update loader.py

* Update vision.py

* Update vision.py

* Update vision.py

* Update loader.py

* is_multimodal

* Update loader.py

* Update loader.py

* Update loader.py

* Update loader.py

* Update vision.py

* Update vision.py

* Update vision.py

* UNSLOTH_DISABLE_STATIC_GENERATION

* Update vision.py

* Auto vision detection

* Sesame

* Whisper

* Update loader.py

* Update loader.py

* Update loader.py

* Update mapper.py

* Update vision.py

* Update vision.py

* Update vision.py

* Update vision.py

* Update vision.py

* Update vision.py

* Update loader.py

* Update loader.py

* Update loader.py

* Update loader.py

* Update _utils.py

* Update rl.py

* versioning

* Update rl.py

* Update rl.py

* Update rl.py

* Update rl.py

* Update rl.py

* logging

* Update pyproject.toml

* Update rl.py

---------

Co-authored-by: Jack Shi Wei Lun <87535974+jackswl@users.noreply.github.com>
Co-authored-by: naliazheli <nalia0316@gmail.com>
Co-authored-by: jeromeku <jerome.ku@gmail.com>
Co-authored-by: Michael Han <107991372+shimmyshimmer@users.noreply.github.com>
 2025-05-28 06:15:12 -07:00
Daniel Han
e2f3a6567d Create LICENSE 2025-05-28 03:27:48 -07:00
jeromeku
15bce315ac Llama4 MoE Grouped GEMM (#2639) 
* add llama4 reference layer

* add llama4 reference impl

* formatting
 2025-05-28 03:26:35 -07:00
Premik
3fb131630f Check the skip_prepare_dataset before accessing dataset fields. #2496 (#2633) 2025-05-28 03:23:59 -07:00
Michael Han
e3e90cf6f3 Update README.md 
Better Qwen3 notebook
 2025-05-26 23:44:41 -07:00
Daniel Han
0bf5e2be15 Flash Attention whls 2025-05-26 22:48:46 -07:00
Datta Nimmaturi
16a007a283 Upgrade trl fix (#2544) 
* Update llama.py making set and reset functions in order to properly use autoSequenceClassification

* Update fast_lora.py, added mixed precising pytorch autocasting

* Update llama.py did not included rotary embeddings in the reset functions correctly

* Update rl.py: correct get reward model added as well as the eval step stuff

* Update rl.py removed function that did not need to be patched

* Update llama.py: kept reset functions and made their names generic

* Update fast_lora.py

* Update rl.py, try except

* Update fast_lora.py, removing downcasting stuff

* Update llama.py removed depircate LLamaLinearScalingRotaryEmbedding

* Update rl.py for VLLM RLOO and PPO

* Update rl.py reverted

* Update rl.py with peft cahnges

* Update rl.py, disabling adapters screws inference up

* Update rl.py getting PPO support

* Update rl.py cleanup

* Update rl.py cleaned up not useful commented code

* Update llama.py, enabled new flag, keep padding

* Upgrade trl fix

Signed-off-by: Dattu Sharma <venkatadattasainimmaturi@gmail.com>

* Update rl.py made changes relative to the review

* Revert accidental patch block for non grpo

Signed-off-by: Dattu Sharma <venkatadattasainimmaturi@gmail.com>

* Fixup sampling params issue

* Fix rl.py regex

Signed-off-by: Dattu Sharma <venkatadattasainimmaturi@gmail.com>

* loss type: grpo, drgrpo and bnpo

Signed-off-by: Dattu Sharma <venkatadattasainimmaturi@gmail.com>

* Add trl version check for vllm colocate mode for RL trainers

* Update rl.py

For TRL 0.18.0 (Main branch of TRL at the time because its on 0.17.0) , the SFT trainer for some reason deletes the labels column and unsloth internal loss funcitons need that column for hte claculations so I add it back in like this.

* Update llama.py, merge it to be dattas llama version

* Update rl.py, sft changes to get 0.18.0 to be working

* Update rl_replacements.py, added hidden state stuff

* Update rl_replacements.py

* Update rl_replacements.py

* Update rl_replacements.py, rechanged the accumlated loss

* Fixup num_iterations>1 for grpo

Signed-off-by: datta0 <venkatadattasainimmaturi@gmail.com>

* Update rl_replacements.py

* no unnecessary logits upcast. fix naming

Signed-off-by: datta0 <venkatadattasainimmaturi@gmail.com>

* Update rl_replacements.py returned hidden states from logprobs

* Update rl_replacements.py removed debug logic

* Update rl_replacements.py, should be fine now

* Update rl_replacements.py, should take new args for GRPO trainer

* Update rl_replacements.py, made it compatible with trl 0.15.2

* Update rl_replacements.py, fixed typo in per tokne-Logps

---------

Signed-off-by: Dattu Sharma <venkatadattasainimmaturi@gmail.com>
Signed-off-by: datta0 <venkatadattasainimmaturi@gmail.com>
Co-authored-by: pluesclues <136766175+pluesclues@users.noreply.github.com>
 2025-05-26 17:20:57 -07:00
Daniel Han
7a3df703a0 Colocate vLLM 2025-05-26 00:37:04 -07:00
Michael Han
2b419ce039 Update README.md 2025-05-25 03:35:43 -07:00
Quentin Gallouédec
a09f93c921 Remove dataset_text_field from SFTConfig (#2609) 2025-05-25 03:20:16 -07:00
Richi
86c77d40bb add: path checking for failed llama cpp builds (#2603) 2025-05-25 03:18:07 -07:00
Daniel Han
911a1def95 Devstral, MedGemma 2025-05-21 07:35:36 -07:00