ceos/unsloth
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions
4,672 commits 391 branches 58 tags 304 MiB
Commit graph

4,672 commits

This branch
This branch
All branches
Author SHA1 Message Date
Daniel Han
d0ce49732a Update rl_replacements.py 2025-07-29 01:59:08 -07:00
Daniel Han
121785dd23 Update rl.py 2025-07-29 01:32:39 -07:00
Daniel Han
b9bcce89e9 Update pyproject.toml 2025-07-29 01:03:19 -07:00
Daniel Han
461ff94bcc Merge branch 'main' of https://github.com/unslothai/unsloth 2025-07-29 00:43:52 -07:00
Daniel Han
c2540a1897 Merge branch 'pr/3055' 2025-07-29 00:43:19 -07:00
Etherll
b14841facf Add gemma-3n chat template to chat_templates.py (#3051) 
* Update chat_templates.py

* Update chat_templates.py
 2025-07-29 00:39:38 -07:00
Daniel Han
4968d3d059 Merge branch 'pr/3052' 2025-07-29 00:38:50 -07:00
Daniel Han
284f7ed538 Fix TRL 0.20.0 2025-07-29 00:36:25 -07:00
Daniel Han
96b1b5ac2b Merge branch 'main' of https://github.com/unslothai/unsloth 2025-07-29 00:35:41 -07:00
Sekinal
5899d5441b Fixed wrong syntax in f-string for exception 2025-07-28 23:16:05 -06:00
Sekinal
55fc4a53f7 Fix: Added specific check for Gemma so models like BERT properly initialize 2025-07-28 22:47:58 -06:00
Etherll
4c3139c069 Update loader.py 2025-07-28 22:15:23 +03:00
Etherll
c2b06fc279 Update _utils.py 2025-07-28 21:48:36 +03:00
Etherll
49d75674f4 Update vision.py 2025-07-28 20:55:07 +03:00
Etherll
b98948db85 Update loader.py 2025-07-28 20:53:38 +03:00
Daniel Han
e812382eaf Update _utils.py 2025-07-28 08:28:18 -07:00
Datta Nimmaturi
9deeaeebeb Fixup multi GPU workload. (#3049) 
* sync all instead

* sync after move and rope init instead

* sync after rope inside

* Return new tensors and no sync

* Sync only current stream

* Fixup mask for xformers

* sync for prefill only

* clean up
 2025-07-28 03:04:49 -07:00
Daniel Han
aec983ea3f Merge branch 'main' into nightly 2025-07-24 23:40:23 -07:00
Edd
8e994f07fb Fix Llama and Gemma inference (#3034) 
* Fix Llama and Gemma inference

* Add simple quality life for CUDA link error (which is not captured since we bypass all error)
 2025-07-24 23:38:20 -07:00
Daniel Han
ffbd2deaea Merge branch 'main' into nightly 2025-07-23 06:08:35 -07:00
Daniel Han
aa391aef66 Update _utils.py 2025-07-23 06:08:31 -07:00
Daniel Han
cc8fe6908b Merge branch 'main' into nightly 2025-07-23 06:04:39 -07:00
Daniel Han
a0836ffdaf Update pyproject.toml 2025-07-23 06:04:35 -07:00
Daniel Han
08f577854f Merge branch 'main' into nightly 2025-07-23 05:53:09 -07:00
Daniel Han
d27e4e44d1 Fix torch compile issues (#3028) 
* 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

* Update rl_replacements.py

* Update rl_replacements.py

* Update rl.py

* Update rl_replacements.py

* Update rl_replacements.py

* logits / temperature

* Update rl_replacements.py

* Update pyproject.toml

* Update rl_replacements.py

* Update rl_replacements.py

* Debugging only

* Update llama.py

* Update llama.py

* Update rl_replacements.py

* Update rl_replacements.py

* Update rl_replacements.py

* Update rl_replacements.py

* Update rl_replacements.py

* Generic efficient GRPO

* Update rl_replacements.py

* Update rl_replacements.py

* Remove debugging

* Update rl_replacements.py

* Update rl_replacements.py

* Update vision.py

* Update llama.py

* Update rl_replacements.py

* versioning

* Update _utils.py

* Update vision.py

* Update mapper.py

* Update loader.py

* Update mapper.py

* Update vision.py

* Update loader.py

* Update vision.py

* Update loader.py

* Update _utils.py

* Update vision.py

* gradient checkpointing

* Gemma 3N fixes

* Update loader.py

* Versioning

* Gemma 3N fixes

* Update vision.py

* Update vision.py

* Update loader.py

* Update vision.py

* Fix setup.py

* setup.py

* Prints

* Update setup.py

* Update setup.py

* Update setup.py

* Update pyproject.toml

* Update pyproject.toml

* Update pyproject.toml

* Update pyproject.toml

* Update pyproject.toml

* Update pyproject.toml

* Update vision.py

* Update vision.py

* Update pyproject.toml

* Update vision.py

* Update _utils.py

* Update __init__.py

* Update __init__.py

* Small fixes

* Update vision.py

* Update vision.py

* versioning

* Update __init__.py

* Update llama.py

* Update rl.py

* Update rl.py

* Update _utils.py

* Update vision.py

* Update vision.py

* compiler stance

* Update _utils.py

* Update pyproject.toml

* Update pyproject.toml

* Update rl_replacements.py

* Update rl_replacements.py

* Update rl_replacements.py

* Update rl_replacements.py

* Update rl.py

* Update rl_replacements.py

* Update rl_replacements.py

* Update rl_replacements.py

* Update rl_replacements.py

* Update rl_replacements.py

* Update rl_replacements.py

* Update rl_replacements.py

* Revert "Revert "Add Qwen2.5-VL-32B-Instruct mapping to fix quantized model me…" (#2990)

This reverts commit 4021da634a.

* skip_guard_eval_unsafe fix

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update llama.py

* Update llama.py

* Fix `quantization_method`

* versioning

* Update _utils.py

* Update _utils.py

* Update _utils.py

* check stride

* Cleanup

* Update rope_embedding.py

* Update gemma2.py

* Fix `set_stance`
 2025-07-23 05:52:28 -07:00
Daniel Han
fec1b2d5f6 Fix set_stance 2025-07-23 05:19:08 -07:00
Daniel Han
56cc02b230 Update gemma2.py 2025-07-23 03:27:52 -07:00
Daniel Han
a5f26f4f76 Update rope_embedding.py 2025-07-23 03:26:16 -07:00
Daniel Han
11d8e5fe53 Cleanup 2025-07-23 03:23:23 -07:00
Daniel Han
bf8049c1c9 check stride 2025-07-23 02:52:29 -07:00
Daniel Han
8fd8a051a9 Merge branch 'main' into nightly 2025-07-23 02:51:00 -07:00
DoubleMathew
f2ef5bd16b falcon force float32 on sm<75 machines (#3026) 2025-07-22 13:18:42 -07:00
Daniel Han
282ae72862 Fix Gemma 2 (#3024) 
* 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

* Update rl_replacements.py

* Update rl_replacements.py

* Update rl.py

* Update rl_replacements.py

* Update rl_replacements.py

* logits / temperature

* Update rl_replacements.py

* Update pyproject.toml

* Update rl_replacements.py

* Update rl_replacements.py

* Debugging only

* Update llama.py

* Update llama.py

* Update rl_replacements.py

* Update rl_replacements.py

* Update rl_replacements.py

* Update rl_replacements.py

* Update rl_replacements.py

* Generic efficient GRPO

* Update rl_replacements.py

* Update rl_replacements.py

* Remove debugging

* Update rl_replacements.py

* Update rl_replacements.py

* Update vision.py

* Update llama.py

* Update rl_replacements.py

* versioning

* Update _utils.py

* Update vision.py

* Update mapper.py

* Update loader.py

* Update mapper.py

* Update vision.py

* Update loader.py

* Update vision.py

* Update loader.py

* Update _utils.py

* Update vision.py

* gradient checkpointing

* Gemma 3N fixes

* Update loader.py

* Versioning

* Gemma 3N fixes

* Update vision.py

* Update vision.py

* Update loader.py

* Update vision.py

* Fix setup.py

* setup.py

* Prints

* Update setup.py

* Update setup.py

* Update setup.py

* Update pyproject.toml

* Update pyproject.toml

* Update pyproject.toml

* Update pyproject.toml

* Update pyproject.toml

* Update pyproject.toml

* Update vision.py

* Update vision.py

* Update pyproject.toml

* Update vision.py

* Update _utils.py

* Update __init__.py

* Update __init__.py

* Small fixes

* Update vision.py

* Update vision.py

* versioning

* Update __init__.py

* Update llama.py

* Update rl.py

* Update rl.py

* Update _utils.py

* Update vision.py

* Update vision.py

* compiler stance

* Update _utils.py

* Update pyproject.toml

* Update pyproject.toml

* Update rl_replacements.py

* Update rl_replacements.py

* Update rl_replacements.py

* Update rl_replacements.py

* Update rl.py

* Update rl_replacements.py

* Update rl_replacements.py

* Update rl_replacements.py

* Update rl_replacements.py

* Update rl_replacements.py

* Update rl_replacements.py

* Update rl_replacements.py

* Revert "Revert "Add Qwen2.5-VL-32B-Instruct mapping to fix quantized model me…" (#2990)

This reverts commit 4021da634a.

* skip_guard_eval_unsafe fix

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update llama.py

* Update llama.py

* Fix `quantization_method`

* versioning

* Update _utils.py

* Update _utils.py

* Update _utils.py
 2025-07-22 04:43:43 -07:00
Daniel Han
e402cc2a74 Update _utils.py 2025-07-22 04:42:51 -07:00
Daniel Han
d23a96b060 Update _utils.py 2025-07-22 04:42:06 -07:00
Daniel Han
9884e991c0 Update _utils.py 2025-07-22 04:39:04 -07:00
Daniel Han
5906e612a0 Merge branch 'main' into nightly 2025-07-22 03:34:55 -07:00
Daniel Han
35dada2a99 Update llama.py 2025-07-22 03:34:51 -07:00
Lei Zhenyuan
7123849cf8 [intel] add for intel path for llama.py (#3012) 
* fix for intel path

* remove unuse code

* Update unsloth/models/llama.py

---------

Co-authored-by: Daniel Han <danielhanchen@gmail.com>
 2025-07-22 03:33:26 -07:00
Lei Zhenyuan
f3e41d0b1e fix for casual mask (#3011) 2025-07-22 03:27:57 -07:00
Daniel Han
673476393b Merge branch 'main' into nightly 2025-07-21 05:35:09 -07:00
Daniel Han
80e7af5b9f Bug fixes (#3017) 
* 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

* Update rl_replacements.py

* Update rl_replacements.py

* Update rl.py

* Update rl_replacements.py

* Update rl_replacements.py

* logits / temperature

* Update rl_replacements.py

* Update pyproject.toml

* Update rl_replacements.py

* Update rl_replacements.py

* Debugging only

* Update llama.py

* Update llama.py

* Update rl_replacements.py

* Update rl_replacements.py

* Update rl_replacements.py

* Update rl_replacements.py

* Update rl_replacements.py

* Generic efficient GRPO

* Update rl_replacements.py

* Update rl_replacements.py

* Remove debugging

* Update rl_replacements.py

* Update rl_replacements.py

* Update vision.py

* Update llama.py

* Update rl_replacements.py

* versioning

* Update _utils.py

* Update vision.py

* Update mapper.py

* Update loader.py

* Update mapper.py

* Update vision.py

* Update loader.py

* Update vision.py

* Update loader.py

* Update _utils.py

* Update vision.py

* gradient checkpointing

* Gemma 3N fixes

* Update loader.py

* Versioning

* Gemma 3N fixes

* Update vision.py

* Update vision.py

* Update loader.py

* Update vision.py

* Fix setup.py

* setup.py

* Prints

* Update setup.py

* Update setup.py

* Update setup.py

* Update pyproject.toml

* Update pyproject.toml

* Update pyproject.toml

* Update pyproject.toml

* Update pyproject.toml

* Update pyproject.toml

* Update vision.py

* Update vision.py

* Update pyproject.toml

* Update vision.py

* Update _utils.py

* Update __init__.py

* Update __init__.py

* Small fixes

* Update vision.py

* Update vision.py

* versioning

* Update __init__.py

* Update llama.py

* Update rl.py

* Update rl.py

* Update _utils.py

* Update vision.py

* Update vision.py

* compiler stance

* Update _utils.py

* Update pyproject.toml

* Update pyproject.toml

* Update rl_replacements.py

* Update rl_replacements.py

* Update rl_replacements.py

* Update rl_replacements.py

* Update rl.py

* Update rl_replacements.py

* Update rl_replacements.py

* Update rl_replacements.py

* Update rl_replacements.py

* Update rl_replacements.py

* Update rl_replacements.py

* Update rl_replacements.py

* Revert "Revert "Add Qwen2.5-VL-32B-Instruct mapping to fix quantized model me…" (#2990)

This reverts commit 4021da634a.

* skip_guard_eval_unsafe fix

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update synthetic.py

* Update llama.py

* Update llama.py

* Fix `quantization_method`

* versioning
 2025-07-21 05:30:14 -07:00
Daniel Han
7e07941383 versioning 2025-07-21 05:15:48 -07:00
Daniel Han
06f1b961f6 Fix quantization_method 2025-07-21 05:14:17 -07:00
Daniel Han
bef2b47599 Update llama.py 2025-07-20 03:19:10 -07:00
Daniel Han
48296987a3 Update llama.py 2025-07-20 03:17:02 -07:00
Daniel Han
9c6b199716 Update synthetic.py 2025-07-20 00:10:41 -07:00
Daniel Han
27503af0dc Update synthetic.py 2025-07-19 03:36:53 -07:00
Daniel Han
75f615891a Update synthetic.py 2025-07-19 03:18:18 -07:00
Daniel Han
6a65ee478c Update synthetic.py 2025-07-19 03:11:33 -07:00