Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

GPTQ dont work anymore #6563

Open
1 task done
kalle07 opened this issue Dec 7, 2024 · 1 comment
Open
1 task done

GPTQ dont work anymore #6563

kalle07 opened this issue Dec 7, 2024 · 1 comment
Labels
bug Something isn't working

Comments

@kalle07
Copy link

kalle07 commented Dec 7, 2024

Describe the bug

see log

Is there an existing issue for this?

  • I have searched the existing issues

Reproduction

10 gptq worked last year tody they dont

gguf work

Screenshot

No response

Logs

22:19:21-672140 INFO     Loading "TheBloke_Luban-Marcoroni-13B-v3-GPTQ"
22:19:21-682114 ERROR    Failed to load the model.
Traceback (most recent call last):
  File "e:\text-generation-webui\modules\ui_model_menu.py", line 232, in load_model_wrapper
    shared.model, shared.tokenizer = load_model(selected_model, loader)
                                     ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "e:\text-generation-webui\modules\models.py", line 93, in load_model
    output = load_func_map[loader](model_name)
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "e:\text-generation-webui\modules\models.py", line 313, in ExLlamav2_HF_loader
    from modules.exllamav2_hf import Exllamav2HF
  File "e:\text-generation-webui\modules\exllamav2_hf.py", line 7, in <module>
    from exllamav2 import (
  File "e:\text-generation-webui\installer_files\env\Lib\site-packages\exllamav2\__init__.py", line 3, in <module>
    from exllamav2.model import ExLlamaV2
  File "e:\text-generation-webui\installer_files\env\Lib\site-packages\exllamav2\model.py", line 35, in <module>
    from exllamav2.config import ExLlamaV2Config
  File "e:\text-generation-webui\installer_files\env\Lib\site-packages\exllamav2\config.py", line 5, in <module>
    from exllamav2.stloader import STFile, cleanup_stfiles
  File "e:\text-generation-webui\installer_files\env\Lib\site-packages\exllamav2\stloader.py", line 5, in <module>
    from exllamav2.ext import none_tensor, exllamav2_ext as ext_c
  File "e:\text-generation-webui\installer_files\env\Lib\site-packages\exllamav2\ext.py", line 291, in <module>
    ext_c = exllamav2_ext
            ^^^^^^^^^^^^^
NameError: name 'exllamav2_ext' is not defined

System Info

win10
update webui 10/24

same with

22:24:22-231318 INFO Loading "TheBloke_Lelantos-Maid-DPO-7B-GPTQ"
22:24:22-241559 ERROR Failed to load the model.
Traceback (most recent call last):
File "e:\text-generation-webui\modules\ui_model_menu.py", line 232, in load_model_wrapper
shared.model, shared.tokenizer = load_model(selected_model, loader)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "e:\text-generation-webui\modules\models.py", line 93, in load_model
output = load_func_maploader
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "e:\text-generation-webui\modules\models.py", line 320, in AutoGPTQ_loader
import modules.AutoGPTQ_loader
File "e:\text-generation-webui\modules\AutoGPTQ_loader.py", line 4, in
from auto_gptq import AutoGPTQForCausalLM, BaseQuantizeConfig
File "e:\text-generation-webui\installer_files\env\Lib\site-packages\auto_gptq_init_.py", line 1, in
from .modeling import AutoGPTQForCausalLM, BaseQuantizeConfig
File "e:\text-generation-webui\installer_files\env\Lib\site-packages\auto_gptq\modeling_init_.py", line 2, in
from .auto import GPTQ_CAUSAL_LM_MODEL_MAP, AutoGPTQForCausalLM
File "e:\text-generation-webui\installer_files\env\Lib\site-packages\auto_gptq\modeling\auto.py", line 9, in
from .decilm import DeciLMGPTQForCausalLM
File "e:\text-generation-webui\installer_files\env\Lib\site-packages\auto_gptq\modeling\decilm.py", line 8, in
from ..nn_modules.fused_llama_attn import FusedLlamaAttentionForQuantizedModel
File "e:\text-generation-webui\installer_files\env\Lib\site-packages\auto_gptq\nn_modules\fused_llama_attn.py", line 6, in
from transformers.models.llama.modeling_llama import (
File "e:\text-generation-webui\installer_files\env\Lib\site-packages\transformers\models\llama\modeling_llama.py", line 32, in
from ...modeling_flash_attention_utils import flash_attention_forward
File "e:\text-generation-webui\installer_files\env\Lib\site-packages\transformers\modeling_flash_attention_utils.py", line 27, in
from flash_attn.bert_padding import index_first_axis, pad_input, unpad_input # noqa
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "e:\text-generation-webui\installer_files\env\Lib\site-packages\flash_attn_init
.py", line 3, in
from flash_attn.flash_attn_interface import (
File "e:\text-generation-webui\installer_files\env\Lib\site-packages\flash_attn\flash_attn_interface.py", line 10, in
import flash_attn_2_cuda as flash_attn_cuda
ImportError: DLL load failed while importing flash_attn_2_cuda: Die angegebene Prozedur wurde nicht gefunden.

@kalle07 kalle07 added the bug Something isn't working label Dec 7, 2024
@SobakaFox
Copy link

SobakaFox commented Dec 12, 2024

Ошибка NameError: name 'exllamav2_ext' is not defined указывает на проблему с инициализацией модуля exllamav2_ext в пакете exllamav2. Эта ошибка обычно возникает из-за отсутствия компиляции или неправильной установки расширений, необходимых для работы ExLlamaV2

pip uninstall flash-attn
pip install flash-attn --no-cache-dir

!!!!ПОСЛЕ КОМАНДЫ pip install flash-attn --no-cache-dir НАЧНЕТСЯ КОМПИЛЯЦИЯ ФАЙЛОВ!!!! компиляция может занять до 2 часов, зависит от вашей системы, не пугайтесь большой нагрузки на RAM и процессор, компиляция это ресурсозатратный процесс. После завершения повторите свою попытку запуска, если ничего не работает следуйте инструкции ниже.

pip install --upgrade exllamav2

Перейдите в директорию exllamav2/ext ----> python setup.py install

pip install torch --upgrade
pip install torchvision --upgrade

проверка версии CUDA на соответствие с версией PyTorch ---> python -m torch.utils.collect_env

если ничего не помогло

pip uninstall exllamav2
pip install exllamav2

удачи

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
bug Something isn't working
Projects
None yet
Development

No branches or pull requests

2 participants