Skip to content

[FIX] qwen vl quantize #2075

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 3 commits into from
Oct 20, 2025
Merged

[FIX] qwen vl quantize #2075

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
31bd632
fix test_qwen2_vl and test_qwen2_5_vl
ZX-ModelCloud Oct 20, 2025
df81ec3
Merge remote-tracking branch 'origin/main' into zx_fix_qwen_vl_quantize
ZX-ModelCloud Oct 20, 2025
ea45e68
update test_qwen2_5_vl acc score
ZX-ModelCloud Oct 20, 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
2 changes: 2 additions & 0 deletions gptqmodel/models/auto.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -187,7 +187,9 @@
"qwen3_moe": Qwen3MoeQModel,
"qwen3_next": Qwen3NextGPTQ,
"qwen2_vl": Qwen2VLQModel,
"qwen2_vl_text": Qwen2VLQModel,
"qwen2_5_vl": Qwen2_5_VLQModel,
"qwen2_5_vl_text": Qwen2_5_VLQModel,
"qwen2_5_omni": Qwen2_5_OmniGPTQ,
"qwen3_omni_moe": Qwen3OmniMoeGPTQ,
"dbrx": DbrxQModel,
Expand Down
4 changes: 2 additions & 2 deletions tests/models/test_qwen2_5_vl.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -13,8 +13,8 @@ class TestQwen2_VL(ModelTest):
NATIVE_MODEL_ID = "/monster/data/model/Qwen2.5-VL-3B-Instruct"
EVAL_TASKS = {
EVAL.LM_EVAL.ARC_CHALLENGE: {
"acc": {"value": 0.2329, "floor_pct": 0.2},
"acc_norm": {"value": 0.2765, "floor_pct": 0.2},
"acc": {"value": 0.4309, "floor_pct": 0.2},
"acc_norm": {"value": 0.4113, "floor_pct": 0.2},
},
}
TRUST_REMOTE_CODE = False
Expand Down
4 changes: 2 additions & 2 deletions tests/models/test_qwen2_vl.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -13,8 +13,8 @@ class TestQwen2_VL(ModelTest):
NATIVE_MODEL_ID = "/monster/data/model/Qwen2-VL-2B-Instruct"
EVAL_TASKS = {
EVAL.LM_EVAL.ARC_CHALLENGE: {
"acc": {"value": 0.2329, "floor_pct": 0.2},
"acc_norm": {"value": 0.2765, "floor_pct": 0.2},
"acc": {"value": 0.3524, "floor_pct": 0.2},
"acc_norm": {"value": 0.3763, "floor_pct": 0.2},
},
}
TRUST_REMOTE_CODE = False
Expand Down
4 changes: 2 additions & 2 deletions tests/models/test_starcode2.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -9,8 +9,8 @@

class TestStarCode2(ModelTest):
NATIVE_MODEL_ID = "/monster/data/model/starcoder2-3b"
NATIVE_ARC_CHALLENGE_ACC = 0.2901
NATIVE_ARC_CHALLENGE_ACC_NORM = 0.3080
NATIVE_ARC_CHALLENGE_ACC = 0.2329
NATIVE_ARC_CHALLENGE_ACC_NORM = 0.2824
TORCH_DTYPE = torch.float16
def test_starcode2(self):
self.quant_lm_eval()
Expand Down