Skip to content

[None][fix] fix config loading for DeepSeek-V3.2 in trtllm-bench #8729

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
lfr-0531 merged 8 commits into from
Oct 29, 2025
+76 −86
Merged

[None][fix] fix config loading for DeepSeek-V3.2 in trtllm-bench #8729

Show file tree
Hide file tree
Changes from 1 commit
Commits
Show all changes
8 commits
Select commit Hold shift + click to select a range
7ad921d
fix transformers issue in trtllm-bench.
lfr-0531 Oct 28, 2025
5aff43f
fix format.
lfr-0531 Oct 28, 2025
28838b3
add load_pretrained_config to config_utils.
lfr-0531 Oct 28, 2025
96a3e17
fix sparse_attention_config for ds_v32.
lfr-0531 Oct 28, 2025
6a638aa
fix config load.
lfr-0531 Oct 28, 2025
658d463
fix format.
lfr-0531 Oct 28, 2025
33de023
add a TODO to the config_utils.
lfr-0531 Oct 29, 2025
09b9232
Merge branch 'main' into user/fanrongl/fix_trtlm_bench_for_ds32
lfr-0531 Oct 29, 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
20 changes: 17 additions & 3 deletions tensorrt_llm/bench/build/build.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -86,9 +86,23 @@ def get_model_config(model_name: str, model_path: Path = None) -> ModelConfig:
Raises:
ValueError: When model is not supported.
"""
if is_nemotron_hybrid(
AutoConfig.from_pretrained(model_path or model_name,
trust_remote_code=True)):
# TODO: remove this once the transformers can support all of those models in _CONFIG_REGISTRY
import transformers
from tensorrt_llm._torch.model_config import _CONFIG_REGISTRY
model_name_or_path = model_path or model_name
config_dict, _ = transformers.PretrainedConfig.get_config_dict(
model_name_or_path,
trust_remote_code=True)
model_type = config_dict.get("model_type")
if model_type in _CONFIG_REGISTRY:
config_class = _CONFIG_REGISTRY[model_type]
model_config = config_class.from_pretrained(
model_name_or_path,
trust_remote_code=True)
else:
model_config = AutoConfig.from_pretrained(model_name_or_path,
trust_remote_code=True)
if is_nemotron_hybrid(model_config):
Comment thread
lfr-0531 marked this conversation as resolved.
Outdated
return NemotronHybridConfig.from_hf(model_name, model_path)
return ModelConfig.from_hf(model_name, model_path)

Expand Down
18 changes: 16 additions & 2 deletions tensorrt_llm/bench/build/dataclasses.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -192,9 +192,23 @@ def get_param_count(cls, model_hf_name, hf_model_path):

@classmethod
def from_hf(cls, model_hf_name, hf_model_path):
# TODO: remove this once the transformers can support all of those models in _CONFIG_REGISTRY
import transformers
from tensorrt_llm._torch.model_config import _CONFIG_REGISTRY
model_name_or_path = hf_model_path or model_hf_name
hf_config = AutoConfig.from_pretrained(
model_name_or_path, trust_remote_code=True).to_dict()
config_dict, _ = transformers.PretrainedConfig.get_config_dict(
model_name_or_path,
trust_remote_code=True)
model_type = config_dict.get("model_type")
if model_type in _CONFIG_REGISTRY:
config_class = _CONFIG_REGISTRY[model_type]
model_config = config_class.from_pretrained(
model_name_or_path,
trust_remote_code=True)
else:
model_config = AutoConfig.from_pretrained(model_name_or_path,
trust_remote_code=True)
hf_config = model_config.to_dict()
Comment thread
lfr-0531 marked this conversation as resolved.
Outdated
param_count = cls.get_param_count(model_hf_name, hf_model_path)

return cls(name=model_hf_name, param_count=param_count, **hf_config)
Expand Down
Loading