ceos/unsloth
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions
2,051 commits 391 branches 58 tags 304 MiB
Commit graph

2,051 commits

This branch
This branch
All branches
Author SHA1 Message Date
jeromeku
9a14edcd2f more registry readme updates 2025-03-31 18:34:18 -07:00
jeromeku
b33970525c clear registry when executing individual model registration file 2025-03-31 18:24:15 -07:00
jeromeku
8d393f29c1 make llama registration more specific 2025-03-31 18:12:52 -07:00
jeromeku
e2cfec6339 add registry readme 2025-03-31 18:11:58 -07:00
jeromeku
ecf70d6caa add quant type test 2025-03-31 17:58:44 -07:00
jeromeku
bb66d454e2 remove deprecated registration api 2025-03-31 17:36:42 -07:00
jeromeku
959727a2d2 add model search method 2025-03-31 17:36:19 -07:00
jeromeku
d93120db9d refactor model registration tests for new registry apis 2025-03-31 17:22:26 -07:00
jeromeku
2ff490e23b add global register models 2025-03-31 17:11:35 -07:00
jeromeku
65ea6356e4 refactor naming for mistral and phi 2025-03-31 17:08:11 -07:00
jeromeku
9a276978d2 rename deepseek registration methods 2025-03-31 17:03:05 -07:00
jeromeku
16f644e95d rename model registration methods 2025-03-31 17:01:51 -07:00
jeromeku
5c402c9e82 add mistral small to registry 2025-03-31 15:31:01 -07:00
jeromeku
025e22b666 remove redundant code when constructing model names 2025-03-31 15:06:08 -07:00
jeromeku
7157f3c47c add deepseek distill models 2025-03-31 12:04:57 -07:00
jeromeku
7dec39e3b6 add deepseek distill llama 2025-03-31 11:47:51 -07:00
jeromeku
767044e7f2 add deepseek r1 zero 2025-03-31 11:32:21 -07:00
jeromeku
9f8f78c90b add deepseek r1 base 2025-03-31 11:30:47 -07:00
jeromeku
e2ff538fc5 add deepseek v3 2025-03-31 11:23:22 -07:00
jeromeku
a46811c471 add phi 2025-03-31 10:22:50 -07:00
jeromeku
756af9f35f add gemma3 to registry 2025-03-31 10:10:20 -07:00
jeromeku
2222e5ad58 add QwenQVQ to registry 2025-03-31 09:45:15 -07:00
jeromeku
76a2b62766 separate registration of base and instruct llama3.2 2025-03-31 09:35:11 -07:00
jeromeku
0f0aa0c476 handle quant types per model size 2025-03-31 09:27:43 -07:00
jeromeku
671dd3dc14 add option to include original model in registry 2025-03-31 09:09:34 -07:00
jeromeku
0395604928 add qwen2.5 models to registry 2025-03-31 08:45:53 -07:00
jeromeku
95da062046 fix quant tag mapping 2025-03-30 16:14:33 -07:00
jeromeku
0130265ca8 add llama model registration 2025-03-30 15:05:33 -07:00
jeromeku
6abdb1fef6 remap literal quant types to QuantType Enum 2025-03-30 14:39:57 -07:00
jeromeku
6b4bf12873 quant types -> Enum 2025-03-30 14:37:30 -07:00
jeromeku
3d1249a551 add llama vision 2025-03-30 11:44:52 -07:00
jeromeku
ab7c51b4a5 start registry reog 2025-03-30 11:36:48 -07:00
jeromeku
35e3b48c2b remove deprecated key function 2025-03-30 11:06:59 -07:00
jeromeku
85209602f3 fix llama registration 2025-03-30 11:06:11 -07:00
jeromeku
410c4b4c76 fix dataclass init 2025-03-30 10:58:51 -07:00
jeromeku
4b3df3d214 refactor global model info dicts to dataclasses 2025-03-30 10:43:00 -07:00
jeromeku
f21d61ae5d move hf hub utils to unsloth/utils 2025-03-28 16:54:38 -07:00
jeromeku
6f6a7a5e9b add model registry 2025-03-28 16:49:12 -07:00
Daniel Han
7afd6afe2c Merge branch 'main' into nightly 2025-03-26 05:21:24 -07:00
Daniel Han
2dc3930a53 Bug Fixes (#2197) 
* Update loader.py

* model names

* Gemma 3 chat template

* Bug fixes

* Update vision.py

* Update vision.py

* Update vision.py

* Update vision.py

* Update vision.py

* Update llama.py

* Update llama.py

* Update rl.py

* Update chat_templates.py

* Update chat_templates.py

* Update vision.py

* Update vision.py

* Update vision.py

* Update loader.py

* Update vision.py

* Update vision.py

* Revert

* Update _utils.py

* forced precision

* Autocast

* Update vision.py

* Update vision.py

* Update rl.py

* Update vision.py

* Update vision.py

* Update vision.py

* Update vision.py

* Update vision.py

* Update rl.py

* vLLM fixes

* constexpr

* Update vision.py

* Update vision.py

* Update vision.py

* Update rl.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 _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update save.py

* New models

* Triton windows update (#1976)

* Update pyproject.toml

* Update README.md

* Update RMS LayerNorm implementation, and list compr. change in chat templates (#1974)

* Update RMS LayerNorm implementation with optimizations and testing suite

* perf: optimize list comprehension in get_ollama_eos_tokens

* Update Zoo

* Update llama.py

* Update llama.py

* Update vision.py

* Update vision.py

* Update vision.py

* Update vision.py

* Update vision.py

* Update vision.py

* Update vision.py

* Update vision.py

* Update vision.py

* Update vision.py

* Update vision.py

* Update vision.py

* Update rl_replacements.py

* Update vision.py

* grpo fix

* Update rl_replacements.py

* Update vision.py

* Update rl_replacements.py

* Update vision.py

* Update mapper.py

* Update vision.py

* Update vision.py

* Update loader.py

* Update vision.py

* Update save.py

* Update save.py

* Update save.py

* Update rl.py

* Update _utils.py

* Version

* Update pyproject.toml

* Update llama.py

* Update llama.py

* bug fix #2008 (#2039)

* fix (#2051)

* Update loader.py

* Update pyproject.toml

* Update pyproject.toml

* Update vision.py

* more prints

* Update loader.py

* LoRA 16bit fix

* Update vision.py

* Update vision.py

* Update _utils.py

* Update vision.py

* move forced float32

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* move print

* Update _utils.py

* disable bfloat16

* Fix forced float32

* move float32

* Ensure trust_remote_code propegates down to unsloth_compile_transformers (#2075)

* Update _utils.py

* Show both `peft_error` and `autoconfig_error`, not just `autoconfig_error` (#2080)

When loading a PEFT model fails, only the `autoconfig_error` is shown. Instead of the `peft_error`, which is what really matters when we're trying to load a PEFT adapter, the user will see something like this:

```
RuntimeError: Unrecognized model in my_model. Should have a `model_type` key in its config.json, or contain one of the following strings in its name: albert, align, altclip, ...
```

This PR just changes it so `autoconfig_error` and `peft_error` are both displayed.

* fix error message (#2046)

* Update vision.py

* Update _utils.py

* Update pyproject.toml

* Update __init__.py

* Update __init__.py

* Update vision.py

* Update vision.py

* Update vision.py

* Update vision.py

* Update vision.py

* Update vision.py

* Update vision.py

* Update vision.py

* Update vision.py

* Update rl_replacements.py

* Update rl_replacements.py

* Update rl_replacements.py

* Update rl_replacements.py

* Update vision.py

* Update vision.py

* Update vision.py

* Update vision.py

* Update vision.py

* Update rl_replacements.py

* Update vision.py

* Update rl_replacements.py

* Update vision.py

* Update vision.py

* Update vision.py

* Update vision.py

* Update vision.py

* Update vision.py

* Remove double generate patch

* Update vision.py

* Update vision.py

* Update vision.py

* Update vision.py

* Update vision.py

* Update mapper.py

* Update vision.py

* fix: config.torch_dtype in LlamaModel_fast_forward_inference (#2091)

* fix: config.torch_dtype in LlamaModel_fast_forward_inference

* Update llama.py

* update for consistency

---------

Co-authored-by: Daniel Han <danielhanchen@gmail.com>

* versioning

* Update vision.py

* Update vision.py

* Update vision.py

* Update vision.py

* Update vision.py

* Update vision.py

* 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 _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

---------

Co-authored-by: Akshay Behl <126911424+Captain-T2004@users.noreply.github.com>
Co-authored-by: Nino Risteski <95188570+NinoRisteski@users.noreply.github.com>
Co-authored-by: Mukkesh Ganesh <mukmckenzie@gmail.com>
Co-authored-by: Kareem <81531392+KareemMusleh@users.noreply.github.com>
Co-authored-by: Xander Hawthorne <167850078+CuppaXanax@users.noreply.github.com>
Co-authored-by: Isaac Breen <isaac.breen@icloud.com>
Co-authored-by: lurf21 <93976703+lurf21@users.noreply.github.com>
Co-authored-by: naliazheli <nalia0316@gmail.com>
Co-authored-by: jeromeku <jerome.ku@gmail.com>
 2025-03-26 05:19:48 -07:00
Daniel Han
9bbba3a511 Update rl_replacements.py 2025-03-26 05:18:18 -07:00
Daniel Han
8e6dfed0ec Versioning 2025-03-26 04:29:44 -07:00
Daniel Han
c6588c73b9 Update vision.py 2025-03-26 04:13:43 -07:00
Daniel Han
599bfeb38c check SDPA for Mistral 3, Pixtral 2025-03-26 04:11:27 -07:00
Daniel Han
98d946acad Update mapper.py 2025-03-26 03:54:57 -07:00
Daniel Han
f197aa2b0a Bug fix 2025-03-26 03:50:46 -07:00
Daniel Han
adf8fb9b80 Update vision.py 2025-03-26 03:13:48 -07:00
Daniel Han
b9914974cd Update vision.py 2025-03-25 23:46:35 -07:00
Daniel Han
fc9f708e8d Update vision.py 2025-03-25 23:31:33 -07:00
Daniel Han
59d04bb523 Update vision.py 2025-03-25 23:20:15 -07:00