From ec04aa048bfce1e467aded0b16b2ec3732c2edc2 Mon Sep 17 00:00:00 2001 From: Daniel Han Date: Mon, 23 Dec 2024 22:27:35 -0800 Subject: [PATCH] dependencies --- pyproject.toml | 168 ++++++++++++++++++++++++++-------------- unsloth/models/llama.py | 8 +- 2 files changed, 114 insertions(+), 62 deletions(-) diff --git a/pyproject.toml b/pyproject.toml index 17ad03bf94..af212891a1 100644 --- a/pyproject.toml +++ b/pyproject.toml @@ -33,10 +33,10 @@ exclude = ["images*"] [project.optional-dependencies] huggingface = [ - "unsloth_zoo>=2024.11.8", + "unsloth_zoo>=2024.12.3", "packaging", "tyro", - "transformers>=4.46.1,<=4.46.3", + "transformers>=4.46.1,!=4.47.0", "datasets>=2.16.0", "sentencepiece>=0.2.0", "tqdm", @@ -51,86 +51,110 @@ huggingface = [ "hf_transfer", ] cu118only = [ - "xformers @ https://download.pytorch.org/whl/cu118/xformers-0.0.22.post7%2Bcu118-cp39-cp39-manylinux2014_x86_64.whl ; python_version=='3.9'", - "xformers @ https://download.pytorch.org/whl/cu118/xformers-0.0.22.post7%2Bcu118-cp310-cp310-manylinux2014_x86_64.whl ; python_version=='3.10'", - "xformers @ https://download.pytorch.org/whl/cu118/xformers-0.0.22.post7%2Bcu118-cp311-cp311-manylinux2014_x86_64.whl ; python_version=='3.11'", + "xformers @ https://download.pytorch.org/whl/cu118/xformers-0.0.22.post7%2Bcu118-cp39-cp39-manylinux2014_x86_64.whl ; python_version=='3.9' ; sys_platform=='linux'", + "xformers @ https://download.pytorch.org/whl/cu118/xformers-0.0.22.post7%2Bcu118-cp310-cp310-manylinux2014_x86_64.whl ; python_version=='3.10' ; sys_platform=='linux'", + "xformers @ https://download.pytorch.org/whl/cu118/xformers-0.0.22.post7%2Bcu118-cp311-cp311-manylinux2014_x86_64.whl ; python_version=='3.11' ; sys_platform=='linux'", ] cu121only = [ - "xformers @ https://download.pytorch.org/whl/cu121/xformers-0.0.22.post7-cp39-cp39-manylinux2014_x86_64.whl ; python_version=='3.9'", - "xformers @ https://download.pytorch.org/whl/cu121/xformers-0.0.22.post7-cp310-cp310-manylinux2014_x86_64.whl ; python_version=='3.10'", - "xformers @ https://download.pytorch.org/whl/cu121/xformers-0.0.22.post7-cp311-cp311-manylinux2014_x86_64.whl ; python_version=='3.11'", + "xformers @ https://download.pytorch.org/whl/cu121/xformers-0.0.22.post7-cp39-cp39-manylinux2014_x86_64.whl ; python_version=='3.9' ; sys_platform=='linux'", + "xformers @ https://download.pytorch.org/whl/cu121/xformers-0.0.22.post7-cp310-cp310-manylinux2014_x86_64.whl ; python_version=='3.10' ; sys_platform=='linux'", + "xformers @ https://download.pytorch.org/whl/cu121/xformers-0.0.22.post7-cp311-cp311-manylinux2014_x86_64.whl ; python_version=='3.11' ; sys_platform=='linux'", ] cu118onlytorch211 = [ - "xformers @ https://download.pytorch.org/whl/cu118/xformers-0.0.23%2Bcu118-cp39-cp39-manylinux2014_x86_64.whl ; python_version=='3.9'", - "xformers @ https://download.pytorch.org/whl/cu118/xformers-0.0.23%2Bcu118-cp310-cp310-manylinux2014_x86_64.whl ; python_version=='3.10'", - "xformers @ https://download.pytorch.org/whl/cu118/xformers-0.0.23%2Bcu118-cp311-cp311-manylinux2014_x86_64.whl ; python_version=='3.11'", + "xformers @ https://download.pytorch.org/whl/cu118/xformers-0.0.23%2Bcu118-cp39-cp39-manylinux2014_x86_64.whl ; python_version=='3.9' ; sys_platform=='linux'", + "xformers @ https://download.pytorch.org/whl/cu118/xformers-0.0.23%2Bcu118-cp310-cp310-manylinux2014_x86_64.whl ; python_version=='3.10' ; sys_platform=='linux'", + "xformers @ https://download.pytorch.org/whl/cu118/xformers-0.0.23%2Bcu118-cp311-cp311-manylinux2014_x86_64.whl ; python_version=='3.11' ; sys_platform=='linux'", ] cu121onlytorch211 = [ - "xformers @ https://download.pytorch.org/whl/cu121/xformers-0.0.23-cp39-cp39-manylinux2014_x86_64.whl ; python_version=='3.9'", - "xformers @ https://download.pytorch.org/whl/cu121/xformers-0.0.23-cp310-cp310-manylinux2014_x86_64.whl ; python_version=='3.10'", - "xformers @ https://download.pytorch.org/whl/cu121/xformers-0.0.23-cp311-cp311-manylinux2014_x86_64.whl ; python_version=='3.11'", + "xformers @ https://download.pytorch.org/whl/cu121/xformers-0.0.23-cp39-cp39-manylinux2014_x86_64.whl ; python_version=='3.9' ; sys_platform=='linux'", + "xformers @ https://download.pytorch.org/whl/cu121/xformers-0.0.23-cp310-cp310-manylinux2014_x86_64.whl ; python_version=='3.10' ; sys_platform=='linux'", + "xformers @ https://download.pytorch.org/whl/cu121/xformers-0.0.23-cp311-cp311-manylinux2014_x86_64.whl ; python_version=='3.11' ; sys_platform=='linux'", ] cu118onlytorch212 = [ - "xformers @ https://download.pytorch.org/whl/cu118/xformers-0.0.23.post1%2Bcu118-cp39-cp39-manylinux2014_x86_64.whl ; python_version=='3.9'", - "xformers @ https://download.pytorch.org/whl/cu118/xformers-0.0.23.post1%2Bcu118-cp310-cp310-manylinux2014_x86_64.whl ; python_version=='3.10'", - "xformers @ https://download.pytorch.org/whl/cu118/xformers-0.0.23.post1%2Bcu118-cp311-cp311-manylinux2014_x86_64.whl ; python_version=='3.11'", + "xformers @ https://download.pytorch.org/whl/cu118/xformers-0.0.23.post1%2Bcu118-cp39-cp39-manylinux2014_x86_64.whl ; python_version=='3.9' ; sys_platform=='linux'", + "xformers @ https://download.pytorch.org/whl/cu118/xformers-0.0.23.post1%2Bcu118-cp310-cp310-manylinux2014_x86_64.whl ; python_version=='3.10' ; sys_platform=='linux'", + "xformers @ https://download.pytorch.org/whl/cu118/xformers-0.0.23.post1%2Bcu118-cp311-cp311-manylinux2014_x86_64.whl ; python_version=='3.11' ; sys_platform=='linux'", ] cu121onlytorch212 = [ - "xformers @ https://download.pytorch.org/whl/cu121/xformers-0.0.23.post1-cp39-cp39-manylinux2014_x86_64.whl ; python_version=='3.9'", - "xformers @ https://download.pytorch.org/whl/cu121/xformers-0.0.23.post1-cp310-cp310-manylinux2014_x86_64.whl ; python_version=='3.10'", - "xformers @ https://download.pytorch.org/whl/cu121/xformers-0.0.23.post1-cp311-cp311-manylinux2014_x86_64.whl ; python_version=='3.11'", + "xformers @ https://download.pytorch.org/whl/cu121/xformers-0.0.23.post1-cp39-cp39-manylinux2014_x86_64.whl ; python_version=='3.9' ; sys_platform=='linux'", + "xformers @ https://download.pytorch.org/whl/cu121/xformers-0.0.23.post1-cp310-cp310-manylinux2014_x86_64.whl ; python_version=='3.10' ; sys_platform=='linux'", + "xformers @ https://download.pytorch.org/whl/cu121/xformers-0.0.23.post1-cp311-cp311-manylinux2014_x86_64.whl ; python_version=='3.11' ; sys_platform=='linux'", ] cu118onlytorch220 = [ - "xformers @ https://download.pytorch.org/whl/cu118/xformers-0.0.24%2Bcu118-cp39-cp39-manylinux2014_x86_64.whl ; python_version=='3.9'", - "xformers @ https://download.pytorch.org/whl/cu118/xformers-0.0.24%2Bcu118-cp310-cp310-manylinux2014_x86_64.whl ; python_version=='3.10'", - "xformers @ https://download.pytorch.org/whl/cu118/xformers-0.0.24%2Bcu118-cp311-cp311-manylinux2014_x86_64.whl ; python_version=='3.11'", + "xformers @ https://download.pytorch.org/whl/cu118/xformers-0.0.24%2Bcu118-cp39-cp39-manylinux2014_x86_64.whl ; python_version=='3.9' ; sys_platform=='linux'", + "xformers @ https://download.pytorch.org/whl/cu118/xformers-0.0.24%2Bcu118-cp310-cp310-manylinux2014_x86_64.whl ; python_version=='3.10' ; sys_platform=='linux'", + "xformers @ https://download.pytorch.org/whl/cu118/xformers-0.0.24%2Bcu118-cp311-cp311-manylinux2014_x86_64.whl ; python_version=='3.11' ; sys_platform=='linux'", ] cu121onlytorch220 = [ - "xformers @ https://download.pytorch.org/whl/cu121/xformers-0.0.24-cp39-cp39-manylinux2014_x86_64.whl ; python_version=='3.9'", - "xformers @ https://download.pytorch.org/whl/cu121/xformers-0.0.24-cp310-cp310-manylinux2014_x86_64.whl ; python_version=='3.10'", - "xformers @ https://download.pytorch.org/whl/cu121/xformers-0.0.24-cp311-cp311-manylinux2014_x86_64.whl ; python_version=='3.11'", + "xformers @ https://download.pytorch.org/whl/cu121/xformers-0.0.24-cp39-cp39-manylinux2014_x86_64.whl ; python_version=='3.9' ; sys_platform=='linux'", + "xformers @ https://download.pytorch.org/whl/cu121/xformers-0.0.24-cp310-cp310-manylinux2014_x86_64.whl ; python_version=='3.10' ; sys_platform=='linux'", + "xformers @ https://download.pytorch.org/whl/cu121/xformers-0.0.24-cp311-cp311-manylinux2014_x86_64.whl ; python_version=='3.11' ; sys_platform=='linux'", ] cu118onlytorch230 = [ - "xformers @ https://download.pytorch.org/whl/cu118/xformers-0.0.27%2Bcu118-cp39-cp39-manylinux2014_x86_64.whl ; python_version=='3.9'", - "xformers @ https://download.pytorch.org/whl/cu118/xformers-0.0.27%2Bcu118-cp310-cp310-manylinux2014_x86_64.whl ; python_version=='3.10'", - "xformers @ https://download.pytorch.org/whl/cu118/xformers-0.0.27%2Bcu118-cp311-cp311-manylinux2014_x86_64.whl ; python_version=='3.11'", - "xformers @ https://download.pytorch.org/whl/cu118/xformers-0.0.27%2Bcu118-cp312-cp312-manylinux2014_x86_64.whl ; python_version=='3.12'", + "xformers @ https://download.pytorch.org/whl/cu118/xformers-0.0.27%2Bcu118-cp39-cp39-manylinux2014_x86_64.whl ; python_version=='3.9' ; sys_platform=='linux'", + "xformers @ https://download.pytorch.org/whl/cu118/xformers-0.0.27%2Bcu118-cp310-cp310-manylinux2014_x86_64.whl ; python_version=='3.10' ; sys_platform=='linux'", + "xformers @ https://download.pytorch.org/whl/cu118/xformers-0.0.27%2Bcu118-cp311-cp311-manylinux2014_x86_64.whl ; python_version=='3.11' ; sys_platform=='linux'", + "xformers @ https://download.pytorch.org/whl/cu118/xformers-0.0.27%2Bcu118-cp312-cp312-manylinux2014_x86_64.whl ; python_version=='3.12' ; sys_platform=='linux'", ] cu121onlytorch230 = [ - "xformers @ https://download.pytorch.org/whl/cu121/xformers-0.0.27-cp39-cp39-manylinux2014_x86_64.whl ; python_version=='3.9'", - "xformers @ https://download.pytorch.org/whl/cu121/xformers-0.0.27-cp310-cp310-manylinux2014_x86_64.whl ; python_version=='3.10'", - "xformers @ https://download.pytorch.org/whl/cu121/xformers-0.0.27-cp311-cp311-manylinux2014_x86_64.whl ; python_version=='3.11'", - "xformers @ https://download.pytorch.org/whl/cu121/xformers-0.0.27-cp312-cp312-manylinux2014_x86_64.whl ; python_version=='3.12'", + "xformers @ https://download.pytorch.org/whl/cu121/xformers-0.0.27-cp39-cp39-manylinux2014_x86_64.whl ; python_version=='3.9' ; sys_platform=='linux'", + "xformers @ https://download.pytorch.org/whl/cu121/xformers-0.0.27-cp310-cp310-manylinux2014_x86_64.whl ; python_version=='3.10' ; sys_platform=='linux'", + "xformers @ https://download.pytorch.org/whl/cu121/xformers-0.0.27-cp311-cp311-manylinux2014_x86_64.whl ; python_version=='3.11' ; sys_platform=='linux'", + "xformers @ https://download.pytorch.org/whl/cu121/xformers-0.0.27-cp312-cp312-manylinux2014_x86_64.whl ; python_version=='3.12' ; sys_platform=='linux'", ] cu118onlytorch240 = [ - "xformers @ https://download.pytorch.org/whl/cu118/xformers-0.0.27.post2%2Bcu118-cp39-cp39-manylinux2014_x86_64.whl ; python_version=='3.9'", - "xformers @ https://download.pytorch.org/whl/cu118/xformers-0.0.27.post2%2Bcu118-cp310-cp310-manylinux2014_x86_64.whl ; python_version=='3.10'", - "xformers @ https://download.pytorch.org/whl/cu118/xformers-0.0.27.post2%2Bcu118-cp311-cp311-manylinux2014_x86_64.whl ; python_version=='3.11'", - "xformers @ https://download.pytorch.org/whl/cu118/xformers-0.0.27.post2%2Bcu118-cp312-cp312-manylinux2014_x86_64.whl ; python_version=='3.12'", + "xformers @ https://download.pytorch.org/whl/cu118/xformers-0.0.27.post2%2Bcu118-cp39-cp39-manylinux2014_x86_64.whl ; python_version=='3.9' ; sys_platform=='linux'", + "xformers @ https://download.pytorch.org/whl/cu118/xformers-0.0.27.post2%2Bcu118-cp310-cp310-manylinux2014_x86_64.whl ; python_version=='3.10' ; sys_platform=='linux'", + "xformers @ https://download.pytorch.org/whl/cu118/xformers-0.0.27.post2%2Bcu118-cp311-cp311-manylinux2014_x86_64.whl ; python_version=='3.11' ; sys_platform=='linux'", + "xformers @ https://download.pytorch.org/whl/cu118/xformers-0.0.27.post2%2Bcu118-cp312-cp312-manylinux2014_x86_64.whl ; python_version=='3.12' ; sys_platform=='linux'", ] cu121onlytorch240 = [ - "xformers @ https://download.pytorch.org/whl/cu121/xformers-0.0.28.post1-cp39-cp39-manylinux_2_28_x86_64.whl ; python_version=='3.9'", - "xformers @ https://download.pytorch.org/whl/cu121/xformers-0.0.28.post1-cp310-cp310-manylinux_2_28_x86_64.whl ; python_version=='3.10'", - "xformers @ https://download.pytorch.org/whl/cu121/xformers-0.0.28.post1-cp311-cp311-manylinux_2_28_x86_64.whl ; python_version=='3.11'", - "xformers @ https://download.pytorch.org/whl/cu121/xformers-0.0.28.post1-cp312-cp312-manylinux_2_28_x86_64.whl ; python_version=='3.12'", + "xformers @ https://download.pytorch.org/whl/cu121/xformers-0.0.28.post1-cp39-cp39-manylinux_2_28_x86_64.whl ; python_version=='3.9' ; sys_platform=='linux'", + "xformers @ https://download.pytorch.org/whl/cu121/xformers-0.0.28.post1-cp310-cp310-manylinux_2_28_x86_64.whl ; python_version=='3.10' ; sys_platform=='linux'", + "xformers @ https://download.pytorch.org/whl/cu121/xformers-0.0.28.post1-cp311-cp311-manylinux_2_28_x86_64.whl ; python_version=='3.11' ; sys_platform=='linux'", + "xformers @ https://download.pytorch.org/whl/cu121/xformers-0.0.28.post1-cp312-cp312-manylinux_2_28_x86_64.whl ; python_version=='3.12' ; sys_platform=='linux'", ] cu124onlytorch240 = [ - "xformers @ https://download.pytorch.org/whl/cu124/xformers-0.0.28.post1-cp39-cp39-manylinux_2_28_x86_64.whl ; python_version=='3.9'", - "xformers @ https://download.pytorch.org/whl/cu124/xformers-0.0.28.post1-cp310-cp310-manylinux_2_28_x86_64.whl ; python_version=='3.10'", - "xformers @ https://download.pytorch.org/whl/cu124/xformers-0.0.28.post1-cp311-cp311-manylinux_2_28_x86_64.whl ; python_version=='3.11'", - "xformers @ https://download.pytorch.org/whl/cu124/xformers-0.0.28.post1-cp312-cp312-manylinux_2_28_x86_64.whl ; python_version=='3.12'", + "xformers @ https://download.pytorch.org/whl/cu124/xformers-0.0.28.post1-cp39-cp39-manylinux_2_28_x86_64.whl ; python_version=='3.9' ; sys_platform=='linux'", + "xformers @ https://download.pytorch.org/whl/cu124/xformers-0.0.28.post1-cp310-cp310-manylinux_2_28_x86_64.whl ; python_version=='3.10' ; sys_platform=='linux'", + "xformers @ https://download.pytorch.org/whl/cu124/xformers-0.0.28.post1-cp311-cp311-manylinux_2_28_x86_64.whl ; python_version=='3.11' ; sys_platform=='linux'", + "xformers @ https://download.pytorch.org/whl/cu124/xformers-0.0.28.post1-cp312-cp312-manylinux_2_28_x86_64.whl ; python_version=='3.12' ; sys_platform=='linux'", + "xformers @ https://download.pytorch.org/whl/cu124/xformers-0.0.28.post1-cp39-cp39-win_amd64.whl ; python_version=='3.9' ; sys_platform=='windows'", + "xformers @ https://download.pytorch.org/whl/cu124/xformers-0.0.28.post1-cp310-cp310-win_amd64.whl ; python_version=='3.10' ; sys_platform=='windows'", + "xformers @ https://download.pytorch.org/whl/cu124/xformers-0.0.28.post1-cp311-cp311-win_amd64.whl ; python_version=='3.11' ; sys_platform=='windows'", + "xformers @ https://download.pytorch.org/whl/cu124/xformers-0.0.28.post1-cp312-cp312-win_amd64.whl ; python_version=='3.12' ; sys_platform=='windows'", ] cu121onlytorch250 = [ - "xformers @ https://download.pytorch.org/whl/cu121/xformers-0.0.28.post2-cp39-cp39-manylinux_2_28_x86_64.whl ; python_version=='3.9'", - "xformers @ https://download.pytorch.org/whl/cu121/xformers-0.0.28.post2-cp310-cp310-manylinux_2_28_x86_64.whl ; python_version=='3.10'", - "xformers @ https://download.pytorch.org/whl/cu121/xformers-0.0.28.post2-cp311-cp311-manylinux_2_28_x86_64.whl ; python_version=='3.11'", - "xformers @ https://download.pytorch.org/whl/cu121/xformers-0.0.28.post2-cp312-cp312-manylinux_2_28_x86_64.whl ; python_version=='3.12'", + "xformers @ https://download.pytorch.org/whl/cu121/xformers-0.0.28.post2-cp39-cp39-manylinux_2_28_x86_64.whl ; python_version=='3.9' ; sys_platform=='linux'", + "xformers @ https://download.pytorch.org/whl/cu121/xformers-0.0.28.post2-cp310-cp310-manylinux_2_28_x86_64.whl ; python_version=='3.10' ; sys_platform=='linux'", + "xformers @ https://download.pytorch.org/whl/cu121/xformers-0.0.28.post2-cp311-cp311-manylinux_2_28_x86_64.whl ; python_version=='3.11' ; sys_platform=='linux'", + "xformers @ https://download.pytorch.org/whl/cu121/xformers-0.0.28.post2-cp312-cp312-manylinux_2_28_x86_64.whl ; python_version=='3.12' ; sys_platform=='linux'", ] cu124onlytorch250 = [ - "xformers @ https://download.pytorch.org/whl/cu124/xformers-0.0.28.post2-cp39-cp39-manylinux_2_28_x86_64.whl ; python_version=='3.9'", - "xformers @ https://download.pytorch.org/whl/cu124/xformers-0.0.28.post2-cp310-cp310-manylinux_2_28_x86_64.whl ; python_version=='3.10'", - "xformers @ https://download.pytorch.org/whl/cu124/xformers-0.0.28.post2-cp311-cp311-manylinux_2_28_x86_64.whl ; python_version=='3.11'", - "xformers @ https://download.pytorch.org/whl/cu124/xformers-0.0.28.post2-cp312-cp312-manylinux_2_28_x86_64.whl ; python_version=='3.12'", + "xformers @ https://download.pytorch.org/whl/cu124/xformers-0.0.28.post2-cp39-cp39-manylinux_2_28_x86_64.whl ; python_version=='3.9' ; sys_platform=='linux'", + "xformers @ https://download.pytorch.org/whl/cu124/xformers-0.0.28.post2-cp310-cp310-manylinux_2_28_x86_64.whl ; python_version=='3.10' ; sys_platform=='linux'", + "xformers @ https://download.pytorch.org/whl/cu124/xformers-0.0.28.post2-cp311-cp311-manylinux_2_28_x86_64.whl ; python_version=='3.11' ; sys_platform=='linux'", + "xformers @ https://download.pytorch.org/whl/cu124/xformers-0.0.28.post2-cp312-cp312-manylinux_2_28_x86_64.whl ; python_version=='3.12' ; sys_platform=='linux'", + "xformers @ https://download.pytorch.org/whl/cu124/xformers-0.0.28.post2-cp39-cp39-win_amd64.whl ; python_version=='3.9' ; sys_platform=='windows'", + "xformers @ https://download.pytorch.org/whl/cu124/xformers-0.0.28.post2-cp310-cp310-win_amd64.whl ; python_version=='3.10' ; sys_platform=='windows'", + "xformers @ https://download.pytorch.org/whl/cu124/xformers-0.0.28.post2-cp311-cp311-win_amd64.whl ; python_version=='3.11' ; sys_platform=='windows'", + "xformers @ https://download.pytorch.org/whl/cu124/xformers-0.0.28.post2-cp312-cp312-win_amd64.whl ; python_version=='3.12' ; sys_platform=='windows'", +] +cu121onlytorch251 = [ + "xformers @ https://download.pytorch.org/whl/cu121/xformers-0.0.28.post3-cp39-cp39-manylinux_2_28_x86_64.whl ; python_version=='3.9' ; sys_platform=='linux'", + "xformers @ https://download.pytorch.org/whl/cu121/xformers-0.0.28.post3-cp310-cp310-manylinux_2_28_x86_64.whl ; python_version=='3.10' ; sys_platform=='linux'", + "xformers @ https://download.pytorch.org/whl/cu121/xformers-0.0.28.post3-cp311-cp311-manylinux_2_28_x86_64.whl ; python_version=='3.11' ; sys_platform=='linux'", + "xformers @ https://download.pytorch.org/whl/cu121/xformers-0.0.28.post3-cp312-cp312-manylinux_2_28_x86_64.whl ; python_version=='3.12' ; sys_platform=='linux'", +] +cu124onlytorch251 = [ + "xformers @ https://download.pytorch.org/whl/cu124/xformers-0.0.28.post3-cp39-cp39-manylinux_2_28_x86_64.whl ; python_version=='3.9' ; sys_platform=='linux'", + "xformers @ https://download.pytorch.org/whl/cu124/xformers-0.0.28.post3-cp310-cp310-manylinux_2_28_x86_64.whl ; python_version=='3.10' ; sys_platform=='linux'", + "xformers @ https://download.pytorch.org/whl/cu124/xformers-0.0.28.post3-cp311-cp311-manylinux_2_28_x86_64.whl ; python_version=='3.11' ; sys_platform=='linux'", + "xformers @ https://download.pytorch.org/whl/cu124/xformers-0.0.28.post3-cp312-cp312-manylinux_2_28_x86_64.whl ; python_version=='3.12' ; sys_platform=='linux'", + "xformers @ https://download.pytorch.org/whl/cu124/xformers-0.0.28.post3-cp39-cp39-win_amd64.whl ; python_version=='3.9' ; sys_platform=='windows'", + "xformers @ https://download.pytorch.org/whl/cu124/xformers-0.0.28.post3-cp310-cp310-win_amd64.whl ; python_version=='3.10' ; sys_platform=='windows'", + "xformers @ https://download.pytorch.org/whl/cu124/xformers-0.0.28.post3-cp311-cp311-win_amd64.whl ; python_version=='3.11' ; sys_platform=='windows'", + "xformers @ https://download.pytorch.org/whl/cu124/xformers-0.0.28.post3-cp312-cp312-win_amd64.whl ; python_version=='3.12' ; sys_platform=='windows'", ] cu118 = [ "unsloth[huggingface]", @@ -207,6 +231,16 @@ cu124-torch250 = [ "bitsandbytes>=0.43.3", "unsloth[cu124onlytorch250]", ] +cu121-torch251 = [ + "unsloth[huggingface]", + "bitsandbytes>=0.43.3", + "unsloth[cu121onlytorch251]", +] +cu124-torch251 = [ + "unsloth[huggingface]", + "bitsandbytes>=0.43.3", + "unsloth[cu124onlytorch251]", +] kaggle = [ "unsloth[huggingface]", ] @@ -244,10 +278,10 @@ colab-ampere-torch220 = [ "flash-attn>=2.6.3", ] colab-new = [ - "unsloth_zoo>=2024.11.8", + "unsloth_zoo>=2024.12.3", "packaging", "tyro", - "transformers>=4.46.1,<=4.46.3", + "transformers>=4.46.1,!=4.47.0", "datasets>=2.16.0", "sentencepiece>=0.2.0", "tqdm", @@ -263,8 +297,8 @@ colab-no-deps = [ "accelerate>=0.34.1", "trl>=0.7.9,!=0.9.0,!=0.9.1,!=0.9.2,!=0.9.3", "peft>=0.7.1", - "xformers<0.0.27", - "bitsandbytes>=0.43.3", + "xformers", + "bitsandbytes>=0.46.1", "protobuf<4.0.0", ] colab = [ @@ -380,8 +414,24 @@ cu124-ampere-torch250 = [ "ninja", "flash-attn>=2.6.3", ] +cu121-ampere-torch251 = [ + "unsloth[huggingface]", + "bitsandbytes>=0.43.3", + "unsloth[cu121onlytorch251]", + "packaging", + "ninja", + "flash-attn>=2.6.3", +] +cu124-ampere-torch251 = [ + "unsloth[huggingface]", + "bitsandbytes>=0.43.3", + "unsloth[cu124onlytorch251]", + "packaging", + "ninja", + "flash-attn>=2.6.3", +] [project.urls] homepage = "http://www.unsloth.ai" documentation = "https://github.com/unslothai/unsloth" -repository = "https://github.com/unslothai/unsloth" \ No newline at end of file +repository = "https://github.com/unslothai/unsloth" diff --git a/unsloth/models/llama.py b/unsloth/models/llama.py index f8fb7d945d..c94514966f 100644 --- a/unsloth/models/llama.py +++ b/unsloth/models/llama.py @@ -66,6 +66,8 @@ except: from huggingface_hub.utils._token import get_token pass from triton import __version__ as triton_version +BlockDiagonalCausalMask = xformers.attn_bias.BlockDiagonalCausalMask if xformers is not None else None + def original_apply_qkv(self, X): Q = self.q_proj(X) @@ -330,7 +332,7 @@ pass def LlamaAttention_fast_forward( self, hidden_states: torch.Tensor, - causal_mask: Optional[xformers.attn_bias.BlockDiagonalCausalMask] = None, + causal_mask: Optional[BlockDiagonalCausalMask] = None, attention_mask: Optional[torch.Tensor] = None, position_ids: Optional[torch.LongTensor] = None, past_key_value: Optional[Tuple[torch.Tensor]] = None, @@ -538,7 +540,7 @@ __DTYPE_MAP = { def LlamaModel_fast_forward( self, input_ids: torch.LongTensor, - causal_mask: Optional[xformers.attn_bias.BlockDiagonalCausalMask] = None, + causal_mask: Optional[BlockDiagonalCausalMask] = None, attention_mask: Optional[torch.Tensor] = None, position_ids: Optional[torch.LongTensor] = None, past_key_values: Optional[List[torch.FloatTensor]] = None, @@ -942,7 +944,7 @@ def CausalLM_fast_forward(fast_forward_inference): def _CausalLM_fast_forward( self, input_ids: torch.LongTensor = None, - causal_mask: Optional[xformers.attn_bias.BlockDiagonalCausalMask] = None, + causal_mask: Optional[BlockDiagonalCausalMask] = None, attention_mask: Optional[torch.Tensor] = None, position_ids: Optional[torch.LongTensor] = None, past_key_values: Optional[List[torch.FloatTensor]] = None,