Skip to content

Refine exllamav2 ut #1001

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.

Sign up for GitHub

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

Merged
merged 6 commits into from
Nov 7, 2025
Merged

Refine exllamav2 ut #1001

Changes from all commits
Commits
Show all changes
6 commits
Select commit Hold shift + click to select a range
719e5ab
fp8 exporting bugfix
WeiweiZhang1 Oct 9, 2025
8e8b04f
Merge branch 'main' of https://github.com/intel/auto-round into main
WeiweiZhang1 Oct 16, 2025
57842a1
Merge branch 'main' of https://github.com/intel/auto-round into main
WeiweiZhang1 Nov 6, 2025
c2daa79
refine exllama backend cuda UT
WeiweiZhang1 Nov 6, 2025
742244a
Update export_to_fp8.py
WeiweiZhang1 Nov 6, 2025
058aaf9
[pre-commit.ci] auto fixes from pre-commit.com hooks
pre-commit-ci[bot] Nov 6, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
6 changes: 4 additions & 2 deletions test/test_cuda/test_exllamav2_backend.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -12,7 +12,7 @@

from auto_round import AutoRound, AutoRoundConfig
from auto_round.eval.evaluation import simple_evaluate_user_model
from auto_round.testing_utils import require_autogptq, require_gptqmodel
from auto_round.testing_utils import require_autogptq, require_gptqmodel, require_package_version_ut


class LLMDataLoader:
Expand All @@ -24,7 +24,7 @@ def __iter__(self):
yield torch.ones([1, 10], dtype=torch.long)


class TestAutoRoundMarlinBackend(unittest.TestCase):
class TestAutoRoundexllamaBackend(unittest.TestCase):

@classmethod
def setUpClass(self):
Expand Down Expand Up @@ -99,6 +99,7 @@ def test_gptqmodel_exllmav2_4bits_asym(self):
shutil.rmtree("./saved", ignore_errors=True)

@require_autogptq
@require_package_version_ut("torch", "<2.6.0")
def test_gptq_exllamav2_4bits_sym(self):
model = AutoModelForCausalLM.from_pretrained(self.model_name, torch_dtype="auto", trust_remote_code=True)
tokenizer = AutoTokenizer.from_pretrained(self.model_name, trust_remote_code=True)
Expand Down Expand Up @@ -130,6 +131,7 @@ def test_gptq_exllamav2_4bits_sym(self):
shutil.rmtree(self.save_folder, ignore_errors=True)

@require_autogptq
@require_package_version_ut("torch", "<2.6.0")
def test_gptq_exllamav2_4bits_sym_group_size(self):
for group_size in [-1, 32, 64, 128, 256, 1024]: ## 384, 768 has accuracy issue
print(f"!!!!!!!!!!!!!!!!!{group_size}!!!!!!!!!!!!!!!!!")
Expand Down