fix(security): harden checkpoint and Energon input loading (#5770) #20962

Run time

Learn about OS pricing on GitHub Actions

Job	Run time
pre-flight / pre-flight	14s
configure	5s
Lint check	27s
cicd-compute-build-matrix	3s
cicd-wait-in-queue	0s
cicd-container-build (gcp, us-east4-docker.pkg.dev/nv-projdgxchipp-20260113193621/megatron-bridge...	3m 21s
cicd-container-build (aws, 766267172432.dkr.ecr.us-east-1.amazonaws.com, nemo-ci-aws-gpu-x2)	4m 38s
Launch_Unit_Tests_Diffusion	2m 34s
generate-test-matrix	8s
Launch_Import_Check	43s
Launch_Unit_Tests_Core_1	7m 41s
Launch_Unit_Tests_Core_2	7m 16s
Launch_Unit_Tests_Core_0	7m 28s
generate-gb200-test-matrix	5s
gb200_L0_Launch_data	5m 15s
gb200_L0_Launch_models_qwen3	12m 0s
gb200_L0_Launch_converter	12m 52s
gb200_L0_Launch_recipes_gemma4_vl	5m 58s
gb200_L0_Launch_recipes_gemma	4m 40s
gb200_L0_Launch_models_deepseek	5m 29s
gb200_L0_Launch_models_nemotron_omni	2m 32s
gb200_L0_Launch_recipes_deepseek_fsdp	5m 11s
gb200_L0_Launch_recipes_deepseek	4m 20s
gb200_L0_Launch_utils	1m 38s
gb200_L0_Launch_recipes_perf_config	1m 53s
gb200_L0_Launch_recipes_llama_1b	3m 11s
gb200_L1_Launch_models_qwen25_vl	4m 2s
gb200_L1_Launch_models_kimi_k25_vl	3m 29s
gb200_L1_Launch_models_qwen3_omni	2m 9s
gb200_L1_Launch_models_ministral3	3m 32s
gb200_L1_Launch_recipes_nemotron_omni	4m 53s
gb200_L1_Launch_models_qwen3_vl	5m 10s
gb200_L1_Launch_models_qwen35_vl	3m 21s
gb200_L1_Launch_recipes_llama_3b	3m 17s
gb200_L1_Launch_recipes_gpt_oss	6m 6s
gb200_L1_Launch_models_glm	4m 47s
gb200_L1_Launch_recipes_ministral3	3m 4s
gb200_L1_Launch_models_stepfun	3m 31s
gb200_L1_Launch_recipes_glm_vl	2m 1s
gb200_L1_Launch_models_glm_vl	4m 0s
gb200_L1_Launch_recipes_qwen	3m 0s
gb200_L1_Launch_models_gpt_oss	6m 23s
L1_Launch_models_glm_vl	8m 54s
gb200_L1_Launch_models_gemma3	4m 2s
gb200_L1_Launch_vlm_inference	2m 2s
L1_Launch_recipes_llama_3b	4m 10s
L1_Launch_models_qwen2	4m 50s
gb200_L1_Launch_recipes_qwen_vl	21m 55s
L1_Launch_models_qwen25_vl	7m 18s
L1_Launch_models_qwen35_vl	3m 37s
L1_Launch_recipes_nemotronh	9m 56s
L1_Launch_models_kimi_k25_vl	3m 20s
L1_Launch_models_gemma3	8m 6s
L1_Launch_recipes_gemma	4m 44s
L1_Launch_models_qwen3_vl	6m 10s
L1_Launch_recipes_gpt_oss	5m 56s
L1_Launch_models_llada15	1m 54s
L1_Launch_models_qwen3_omni	2m 33s
L1_Launch_models_glm	5m 59s
gb200_L1_Launch_models_qwen2	3m 43s
L1_Launch_recipes_glm_vl	3m 46s
L1_Launch_recipes_qwen	3m 40s
L1_Launch_recipes_nemotron_omni	3m 46s
L1_Launch_recipes_qwen_vl	23m 50s
L0_Launch_converter_generate	8m 38s
L1_Launch_recipes_ministral3	5m 4s
L0_Launch_megatron_fsdp	3m 25s
L0_Launch_models_deepseek	5m 45s
L0_Launch_recipes_deepseek	3m 1s
L0_Launch_models_gemma4_vl	5m 58s
L0_Launch_recipes_llama_1b	4m 7s
L0_Launch_models_qwen3	10m 0s
L1_Launch_models_stepfun	4m 29s
L0_Launch_training_distributed_save	2m 11s
L0_Launch_data	6m 27s
L1_Launch_vlm_inference	4m 52s
L1_Launch_models_gpt_oss	8m 13s
L0_Launch_training_tensor_inspect	3m 10s
L0_Launch_training_dspark	1m 59s
L0_Launch_training_local_checkpointing	3m 58s
L0_Launch_converter_fsdp	4m 2s
L0_Launch_recipes_gemma4_vl	3m 33s
L0_Launch_models_flux	5m 58s
L0_Launch_models_nemotron_omni	2m 48s
L0_Launch_training_callbacks	3m 21s
L0_Launch_converter	13m 21s
L0_Launch_training_resiliency	4m 2s
L0_Launch_training_megatron_mimo	4m 3s
L0_Launch_training_decentralized_pg	4m 14s
L0_Launch_training_pretrain	5m 16s
L0_Launch_training_finetune	7m 19s
L1_Launch_models_ministral3	5m 50s
L0_Launch_recipes_perf_config	2m 19s
L0_Launch_wan_ckpt_conversion	3m 41s
L0_Launch_utils	2m 2s
matrix.script	-1s
gb200_${{ matrix.script }}	-1s
gb200_${{ matrix.script }}	-1s
matrix.script	-1s
Nemo_CICD_Test	20s
Coverage (e2e)	35s
Coverage (unit-test)	32s
Coverage_Fake	0s
	7h 41m 7s

Navigation Menu

Uh oh!

fix(security): harden checkpoint and Energon input loading (#5770) #20962

Usage

fix(security): harden checkpoint and Energon input loading (#5770) #20962

Uh oh!

Run time