d-flex flex-column width-full">

Keep DBuffer placement validation local

bed9193

Select commit

Loading

Failed to load commit list.

Uh oh!

There was an error while loading. Please reload this page.

DCO

DCO

Auto Swap Labels on: pull_request_target

check-approval

Codeowners Approval Workflow on: push

pre-flight / pre-flight

codeowners-approval

multi-approval-bot-summary

Installation Test on: push

pre-flight / pre-flight

Pip - Python3.12 - AMD64/Linux - NGC PyTorch

UV - Python3.12 - AMD64/Linux - NGC PyTorch

Install test summary

CICD Megatron-LM on: push 520

is-not-external-contributor

merge-queue-notification

DCO

pre-flight / pre-flight

cicd-compute-build-matrix

configure

linting

cicd-wait-in-queue

cicd-parse-downstream-testing

cicd-container-build (aws, 766267172432.dkr.ecr.us-east-1.amazonaws.com, nvidia-ci-aws-gpu-x8)

cicd-container-build (gcp, us-east4-docker.pkg.dev/nv-projdgxchipp-20260113193621/megatron-lm, nv...

cicd-mbridge-testing

cicd-parse-unit-tests

cicd-mbridge-testing-notify

tests/unit_tests/test_model_configs.py - latest

tests/unit_tests/test_fp8_param.py - latest

tests/unit_tests/pipeline_parallel/**/*.py - latest

tests/unit_tests/models/**/*.py - latest

tests/unit_tests/data/**/*.py - latest

tests/unit_tests/dist_checkpointing/test_optimizer.py - latest

tests/unit_tests/dist_checkpointing/**/*.py - latest

tests/unit_tests/dist_checkpointing/models/**/*.py - latest

tests/unit_tests/dist_checkpointing/models/test_moe_experts.py - latest

tests/unit_tests/transformer/**/*.py - latest

tests/unit_tests/inference/engines/test_dynamic_engine.py - latest

tests/unit_tests/inference/test_hybrid_moe.py - latest

tests/unit_tests/inference/**/*.py - latest

tests/unit_tests/rl/**/*.py - latest

tests/unit_tests/post_training/**/*.py - latest

tests/unit_tests/resharding/**/*.py - latest

tests/unit_tests/ssm/test_gated_delta_net.py - latest

tests/unit_tests/ssm/**/*.py - latest

tests/unit_tests/transformer/moe/**/*.py - latest

tests/unit_tests/distributed/megatron_fsdp/**/*.py - latest

tests/unit_tests/a2a_overlap/**/*.py - latest

tests/unit_tests/**/*.py - latest

tests/unit_tests/transformer/test_multi_latent_attention.py - latest

tests/unit_tests/transformer/test_multi_token_prediction.py - latest

cicd-integration-gate

cicd-parse-integration-tests-h100

cicd-parse-integration-tests-gb200

common/moe_perf - latest

gpt/gpt3_mcore_te_tp1_pp1_dist_optimizer_fim_dataset - latest

gpt/gpt3_mcore_te_tp1_pp1_dist_optimizer_no_mmap_bin_files - latest

gpt/gpt3_mcore_te_tp1_pp1_resume_torch_dist_uniform_full_recompute - latest

gpt/gpt3_mcore_te_tp1_pp2_resume_torch_dist_rope_embeddings - latest

gpt/gpt3_mcore_te_tp1_pp2_resume_torch_dist_rope_embeddings_interleaved_no_fusion - latest

gpt/gpt3_mcore_te_tp1_pp4_resume_torch_dist_disable_bias_linear - latest

gpt/gpt3_mcore_te_tp1_pp4_resume_torch_dist_persistent_disable_bias_linear - latest

gpt/gpt3_mcore_te_tp1_pp4_resume_torch_dist_untie_embeddings_and_outputs - latest

gpt/gpt3_mcore_te_tp1_pp4_vp1 - latest

gpt/gpt3_mcore_te_tp1_pp4_vp1_dist_optimizer_overlap_grad_reduce_param_gather_overlap_optimizer - latest

gpt/gpt3_mcore_te_tp1_pp4_vp1_resume_torch_decoupled_lr - latest

gpt/gpt3_mcore_te_tp1_pp4_vp1_resume_torch_dist_calculate_per_token_loss - latest

gpt/gpt3_mcore_te_tp1_pp4_vp1_resume_torch_dist_dist_optimizer_overlap_grad_reduce - latest

gpt/gpt3_mcore_te_tp1_pp4_vp1_uneven_pipeline - latest

gpt/gpt3_mcore_te_tp1_pp4_vp2_account_for_embedding_loss_in_pipeline_split - latest

gpt/gpt3_mcore_te_tp2_pp1_gdn - latest

gpt/gpt3_mcore_te_tp2_pp1_gdn_no_nvrx_async - latest

gpt/gpt3_mcore_te_tp2_pp1_gdn_no_nvrx_async_mcore - latest

gpt/gpt3_mcore_te_tp2_pp1_gdn_no_nvrx_sync - latest

gpt/gpt3_mcore_te_tp2_pp1_modelopt_distill_resume - latest

gpt/gpt3_mcore_te_tp2_pp1_resume_torch_dist_cp2_nondeterministic - latest

gpt/gpt3_mcore_te_tp2_pp1_resume_torch_dist_multi_dist_optimizer_instances - latest

gpt/gpt3_mcore_te_tp2_pp2_cp2 - latest

gpt/gpt3_mcore_te_tp2_pp2_cp2_calculate_per_token_loss - latest

gpt/gpt3_mcore_te_tp2_pp2_cp2_calculate_per_token_loss_nondeterministic - latest

gpt/gpt3_mcore_te_tp2_pp2_cp2_etp4_calculate_per_token_loss_dp_last - latest

gpt/gpt3_mcore_te_tp2_pp2_cp2_etp4_calculate_per_token_loss_nondeterministic_dp_last - latest

gpt/gpt3_mcore_te_tp2_pp2_cp2_etp4_dp_last - latest

gpt/gpt3_mcore_te_tp2_pp2_cp2_etp4_nondeterministic_dp_last - latest

gpt/gpt3_mcore_te_tp2_pp2_cp2_nondeterministic - latest

gpt/gpt3_mcore_te_tp2_pp2_cross_entropy_loss_fusion - latest

gpt/gpt3_mcore_te_tp2_pp2_dsa - latest

gpt/gpt3_mcore_te_tp2_pp2_mla - latest

gpt/gpt3_mcore_te_tp2_pp2_resume_torch_dist - latest

gpt/gpt3_mcore_te_tp2_pp2_resume_torch_dist_cp2_nondeterministic - latest

gpt/gpt3_mcore_te_tp2_pp2_resume_torch_dist_cross_entropy_loss_fusion - latest

gpt/gpt3_mcore_te_tp2_pp2_resume_torch_dist_ddp_average_in_collective - latest

gpt/gpt3_mcore_te_tp2_pp2_resume_torch_dist_defer_embedding_wgrad_compute - latest

gpt/gpt3_mcore_te_tp2_pp2_resume_torch_dist_reshard_1x4xNone - latest

gpt/gpt3_mcore_te_tp2_zp_z3_resume_fsdp_dtensor - latest

gpt/gpt3_mcore_te_tp4_pp1_dist_optimizer_overlap_grad_reduce_param_gather - latest

gpt/gpt3_mcore_te_tp4_pp1_resume_torch_dist_dist_optimizer_overlap_grad_reduce - latest

gpt/gpt3_mcore_te_tp4_pp1_resume_torch_dist_dist_optimizer_overlap_grad_reduce_param_gather - latest

gpt/gpt3_mcore_te_tp4_pp1_resume_torch_dist_qk_layernorm_test_mode - latest

gpt/gpt3_mcore_te_tp4_pp2_resume_torch_dist_reshard_8x1xNone - latest

gpt/gpt3_mcore_tp2_pp2_uninstall_te - latest

gpt/gpt_dynamic_inference_tp2_pp2_dp2_583m_logitsmatch_zmq - latest

gpt/gpt_grpo_basic_function - latest

gpt/gpt_static_inference_tp1_pp1_16b_multiprompt_tokensmatch - latest

gpt/gpt_static_inference_tp1_pp1_583m_cudagraphs - latest

gpt/gpt_static_inference_tp1_pp1_583m_fp8_cudagraphs - latest

gpt/gpt_static_inference_tp1_pp1_583m_logitsmatch - latest

hybrid/hybrid_dynamic_inference_tp1_pp1_dp8_583m - latest

hybrid/hybrid_dynamic_inference_tp1_pp1_dp8_583m_chunked_prefill - latest

hybrid/hybrid_dynamic_inference_tp1_pp1_dp8_583m_flashinfer - latest

hybrid/hybrid_dynamic_inference_tp1_pp1_dp8_583m_mamba_bf16_states - latest

hybrid/hybrid_mr_mcore_te_tp1_pp1_cp1_dgx_a100_1N8G - latest

hybrid/hybrid_mr_mcore_te_tp1_pp2_vpp2_cp1_dgx_a100_1N8G - latest

hybrid/hybrid_mr_mcore_te_tp1_pp4_cp1_dgx_a100_1N8G - latest

hybrid/hybrid_mr_mcore_te_tp2_pp1_cp1_dgx_a100_1N8G - latest

hybrid/hybrid_mr_mcore_te_tp2_pp1_cp4_dgx_a100_1N8G - latest

moe/deepseek_proxy_fsdp_ep2_fsdp2_ep_overlap - latest

moe/gpt3_mcore_te_tp1_pp1_te_4experts_groupedGEMM_op_fuser - latest

moe/gpt3_mcore_te_tp1_pp2_resume_torch_dist_reshard_2x1x4_te_8experts2parallel_dist_optimizer - latest

moe/gpt3_mcore_te_tp2_pp1_resume_torch_dist_te_8experts2parallel_dist_optimizer - latest

moe/gpt3_mcore_te_tp2_pp1_resume_torch_dist_te_8experts2parallel_multi_dist_optimizer_instances - latest

moe/gpt3_mcore_te_tp2_pp1_te_8experts2parallel_ddp_average_in_collective - latest

moe/gpt3_mcore_te_tp2_pp1_te_8experts2parallel_overlap_grad_reduce_param_gather_groupedGEMM - latest

moe/gpt3_mcore_te_tp2_pp1_te_a2a_ovlp_8experts_etp1_ep4 - latest

moe/gpt3_moe_mcore_te_ep8_resume_torch_dist_dist_muon - latest

moe/gpt3_moe_mcore_te_ep8_resume_torch_dist_dist_optimizer - latest

moe/gpt3_moe_mcore_te_ep8_resume_torch_dist_muon - latest

moe/gpt3_moe_mcore_te_tp4_ep2_etp2_pp2_resume_torch_dist_dist_optimizer - latest

moe/gpt3_moe_mcore_te_tp4_ep2_etp2_pp2_scoped_cudagraph - latest

gpt/gpt3_mcore_reruns_persistent_1_1node - latest

gpt/gpt3_mcore_te_tp1_pp1_dist_optimizer_no_mmap_bin_files_1node - latest

gpt/gpt3_mcore_te_tp1_pp1_resume_torch_dist_dist_optimizer_1node - latest

gpt/gpt3_mcore_te_tp1_pp2_resume_torch_dist_rope_embeddings_interleaved_no_fusion_1node - latest

gpt/gpt3_mcore_te_tp1_pp4_resume_torch_dist_persistent_disable_bias_linear_1node - latest

gpt/gpt3_mcore_te_tp1_pp4_resume_torch_dist_untie_embeddings_and_outputs_1node - latest

gpt/gpt3_mcore_te_tp1_pp4_vp1_resume_torch_decoupled_lr_1node - latest

gpt/gpt3_mcore_te_tp1_pp4_vp1_resume_torch_dist_dist_optimizer_overlap_grad_reduce_1node - latest

gpt/gpt3_mcore_te_tp1_pp4_vp1_uneven_pipeline_1node - latest

gpt/gpt3_mcore_te_tp2_pp1_resume_torch_dist_cp2_nondeterministic_1node - latest

gpt/gpt3_mcore_te_tp2_pp2_cp2_1node - latest

gpt/gpt3_mcore_te_tp2_pp2_cp2_calculate_per_token_loss_1node - latest

gpt/gpt3_mcore_te_tp2_pp2_cp2_calculate_per_token_loss_nondeterministic_1node - latest

gpt/gpt3_mcore_te_tp2_pp2_cp2_nondeterministic_1node - latest

gpt/gpt3_mcore_te_tp2_pp2_cross_entropy_loss_fusion_1node - latest

gpt/gpt3_mcore_te_tp2_pp2_mla_1node - latest

gpt/gpt3_mcore_te_tp2_pp2_resume_torch_dist_cp2_nondeterministic_1node - latest

gpt/gpt3_mcore_te_tp2_pp2_resume_torch_dist_defer_embedding_wgrad_compute_1node - latest

gpt/gpt3_mcore_te_tp2_pp2_resume_torch_dist_reshard_1x4xNone_1node - latest

gpt/gpt3_mcore_te_tp2_zp_z3_resume_fsdp_dtensor_1node - latest

gpt/gpt3_mcore_te_tp4_pp1_resume_torch_dist_dist_optimizer_overlap_grad_reduce_param_gather_1node - latest

gpt/gpt3_mcore_te_tp4_pp1_resume_torch_dist_qk_layernorm_test_mode_1node - latest

gpt/gpt3_mcore_te_tp4_pp2_resume_torch_dist_reshard_8x1xNone_1node - latest

moe/deepseek_proxy_fsdp_ep2_fsdp2_1node - latest

moe/gpt3_mcore_te_tp2_pp1_resume_torch_dist_te_8experts2parallel_dist_optimizer_1node - latest

moe/gpt3_mcore_te_tp2_pp1_te_8experts2parallel_ddp_average_in_collective_1node - latest

moe/gpt3_mcore_te_tp2_pp1_te_8experts2parallel_overlap_grad_reduce_param_gather_groupedGEMM_1node - latest

moe/gpt3_moe_mcore_te_ep8_resume_torch_dist_muon_1node - latest

moe/gpt3_moe_mcore_te_tp4_ep2_etp2_pp2_resume_torch_dist_dist_optimizer_1node - latest

Nemo_CICD_Test

Coverage_Fake

Coverage (unit-test)

cleanup-taint-node

Request NVSkills CI on: push

request

Copyright check on: push

pre-flight / pre-flight

copyright-check / main

copyright-check-summary

Build, validate, and release Megatron-Core on: push 9

pre-flight / pre-flight

bump / check-admin-permission

bump / bump-next-version

build-test-publish-wheels / build-and-test-wheels (megatron-core, arm64, quay.io/pypa/manylinux_2_28_aarch64)

build-test-publish-wheels / build-and-test-wheels (megatron-core, amd64, quay.io/pypa/manylinux_2_28_x86_64)

build-test-publish-wheels / build-and-test-wheels (megatron-fsdp, quay.io/pypa/manylinux_2_28_x86_64, amd64)

build-test-publish-wheels / publish-wheels (megatron-core, arm64)

build-test-publish-wheels / publish-wheels (megatron-core, amd64)

build-test-publish-wheels / publish-wheels (megatron-fsdp, amd64)

finalize / create-gh-release

finalize / build-docs / Build docs

finalize / notify

finalize / publish-docs

release-summary

Review Trigger on: pull_request_review

signal

DCO / DCO succeeded Jun 16, 2026 in 1s

DCO

All commits are signed off!

View more details on DCO

Loading

Re-running checks...