Skip to content

[https://nvbugs/6181383][fix] Build inner text/vision/audio sub-configs as empty PretrainedConfig() then setat #14399

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
2ez4bz merged 2 commits into from
Jun 9, 2026
Merged

[https://nvbugs/6181383][fix] Build inner text/vision/audio sub-configs as empty PretrainedConfig() then setat #14399

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
6d70810
[nvbugs/6181383][fix] mistral config loader compat with transformers …
tensorrt-cicd May 21, 2026
f6056ab
[nvbugs/6181383][chore] Remove stale waiver after fix
tensorrt-cicd May 25, 2026
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
5 changes: 3 additions & 2 deletions tensorrt_llm/_torch/models/modeling_deepseekv3.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -340,15 +340,16 @@ def split_kv_b_proj(kv_b_proj: torch.Tensor,
# preserved as `_ckpt_num_nextn_predict_layers`.
ckpt_num_nextn_predict_layers = (
getattr(self.config, '_ckpt_num_nextn_predict_layers', None)
or self.config.num_nextn_predict_layers)
or getattr(self.config, 'num_nextn_predict_layers', None))

def detect_shared_mtp_weights() -> bool:
# Detect if MTP layers share checkpoint weights (model has more MTP
# layer instances than the checkpoint provides). In this case,
# multiple model MTP layers map to the same checkpoint layer via
# modulo, and mark_consumed must be skipped to avoid deleting
# weights that later MTP layers still need.
model_nextn = self.config.num_nextn_predict_layers or 0
model_nextn = getattr(self.config, 'num_nextn_predict_layers',
None) or 0
return model_nextn > (ckpt_num_nextn_predict_layers or 0) > 0

has_shared_mtp_weights = detect_shared_mtp_weights()
Expand Down
1 change: 0 additions & 1 deletion tests/integration/test_lists/waives.txt
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -140,7 +140,6 @@ accuracy/test_llm_api_pytorch.py::TestStep3_7::test_fp8_block_scales[tp_size=4-e
accuracy/test_llm_api_pytorch_multimodal.py::TestExaone4_5_33B::test_auto_dtype[forced_chunked_prefill] SKIP (https://nvbugs/6211189)
accuracy/test_llm_api_pytorch_multimodal.py::TestExaone4_5_33B::test_auto_dtype[full_budget] SKIP (https://nvbugs/6211189)
accuracy/test_llm_api_pytorch_multimodal.py::TestGemma3_27BInstruct::test_fp8_prequantized SKIP (https://nvbugs/6215689)
accuracy/test_llm_api_pytorch_multimodal.py::TestMistralLarge3_675B::test_nvfp4_4gpus[latency_moe_trtllm] SKIP (https://nvbugs/6181383)
accuracy/test_llm_api_pytorch_multimodal.py::TestStep3_7::test_fp8_block_scales SKIP (https://nvbugs/6274932)
accuracy/test_llm_api_pytorch_ray.py::TestLlama3_1_8BInstruct::test_pp2_ray SKIP (https://nvbugs/6094070)
cpp/test_e2e.py::test_benchmarks[bart-90] SKIP (https://nvbugs/5550689)
Expand Down
Loading