Skip to content

ci: Bump Megatron-Bridge to e9529c3 #2835

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

Open
svcnvidia-nemo-ci wants to merge 1 commit into
base: main
Choose a base branch
from
+1 −1
Open

ci: Bump Megatron-Bridge to e9529c3 #2835

Changes from all commits
Commits
Show all changes
1 commit
Select commit
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: 1 addition & 1 deletion 3rdparty/Megatron-Bridge-workspace/Megatron-Bridge
Open in desktop
Submodule Megatron-Bridge updated 62 files
+1 −1 .main.commit
+1 −1 3rdparty/Megatron-LM
+102 −1 docs/performance-summary-archive.md
+69 −75 docs/performance-summary.md
+24 −3 examples/conversion/adapter/README.md
+188 −7 examples/conversion/adapter/export_adapter.py
+13 −3 examples/conversion/compare_hf_and_megatron/compare.py
+0 −1 examples/conversion/convert_checkpoints_multi_gpu.py
+93 −28 examples/models/deepseek_v4/conversion.sh
+18 −10 examples/models/nemotron/nemotron_vl/finetune_nemotron_nano_v2_vl.py
+7 −2 examples/peft/merge_lora.py
+2 −0 pyproject.toml
+24 −15 scripts/inference/text_generation.py
+13 −21 skills/nemo-mbridge-perf-activation-recompute/BENCHMARK.md
+2 −0 skills/nemo-mbridge-perf-activation-recompute/SKILL.md
+11 −11 skills/nemo-mbridge-perf-activation-recompute/skill-card.md
+1 −1 skills/nemo-mbridge-perf-activation-recompute/skill.oms.sig
+11 −17 skills/nemo-mbridge-perf-cuda-graphs/BENCHMARK.md
+2 −0 skills/nemo-mbridge-perf-cuda-graphs/SKILL.md
+14 −8 skills/nemo-mbridge-perf-cuda-graphs/skill-card.md
+1 −1 skills/nemo-mbridge-perf-cuda-graphs/skill.oms.sig
+42 −11 src/megatron/bridge/models/conversion/auto_bridge.py
+3 −0 src/megatron/bridge/models/conversion/model_bridge.py
+11 −2 src/megatron/bridge/models/conversion/peft_bridge.py
+11 −2 src/megatron/bridge/models/conversion/utils.py
+2 −1 src/megatron/bridge/models/ernie_vl/modeling_ernie45_vl/model.py
+24 −2 src/megatron/bridge/models/falcon_h1/modeling_falconh1/falconh1_layer.py
+6 −2 src/megatron/bridge/models/falcon_h1/modeling_falconh1/falconh1_model.py
+2 −1 src/megatron/bridge/models/gemma_vl/modeling_gemma3_vl.py
+2 −1 src/megatron/bridge/models/gemma_vl/modeling_gemma4_vl.py
+2 −1 src/megatron/bridge/models/glm_vl/modeling_glm_45v.py
+2 −1 src/megatron/bridge/models/kimi_vl/modeling_kimi_k25_vl.py
+2 −1 src/megatron/bridge/models/ministral3/modeling_ministral3.py
+6 −0 src/megatron/bridge/models/qwen3_asr/hf_qwen3_asr/configuration_qwen3_asr.py
+3 −1 src/megatron/bridge/models/qwen3_asr/modeling_qwen3_asr/thinker_model.py
+3 −1 src/megatron/bridge/models/qwen_omni/modeling_qwen25_omni/thinker_model.py
+3 −1 src/megatron/bridge/models/qwen_omni/modeling_qwen3_omni/thinker_model.py
+2 −1 src/megatron/bridge/models/qwen_vl/modeling_qwen25_vl.py
+3 −1 src/megatron/bridge/models/qwen_vl/modelling_qwen3_vl/model.py
+1 −1 src/megatron/bridge/models/qwen_vl/modelling_qwen3_vl/text_model.py
+2 −0 src/megatron/bridge/models/stepfun/step35_provider.py
+33 −5 src/megatron/bridge/peft/utils.py
+107 −12 src/megatron/bridge/recipes/nemotron_vl/nemotron_nano_v2_vl.py
+68 −10 tests/functional_tests/test_groups/training/test_seqpacking_cp_example.py
+161 −0 tests/unit_tests/examples/test_convert_checkpoints_multi_gpu.py
+36 −0 tests/unit_tests/examples/test_nemotron_vl_finetune.py
+66 −0 tests/unit_tests/models/falcon_h1/test_falcon_h1_layers.py
+74 −0 tests/unit_tests/models/qwen3_asr/test_qwen3_asr_config.py
+4 −2 tests/unit_tests/models/qwen_omni/modeling_qwen3_omni/test_omni_model.py
+50 −0 tests/unit_tests/models/qwen_vl/modelling_qwen3_vl/test_text_model_forward.py
+54 −12 tests/unit_tests/models/stepfun/test_step35_provider.py
+514 −5 tests/unit_tests/models/test_adapter_export.py
+1 −0 tests/unit_tests/models/test_auto_bridge.py
+38 −8 tests/unit_tests/models/test_model_bridge_lora.py
+71 −0 tests/unit_tests/models/test_sequence_parallel_scatter_groups.py
+48 −0 tests/unit_tests/peft/test_utils.py
+125 −3 tests/unit_tests/recipes/test_nemotron_vl_recipes.py
+32 −0 tests/unit_tests/scripts/test_text_generation.py
+15 −0 tests/unit_tests/test_compare_mask_handling.py
+ − tutorials/megatron_mimo/figures/qwen35_vl_loss_parity_400iter.png
+53 −29 tutorials/megatron_mimo/qwen35-vl-non-colocated-sft.md
+40 −38 uv.lock
Loading