Skip to navigation

Training Report Schema

View as Markdown

training_report.json is the unified training output for NeMoRun, MegatronRun, and Megatron-Bridge.

Versioning

The top-level schema_version field uses MAJOR.MINOR versioning:

  • Increment MAJOR when removing, renaming, or incompatibly changing a field.
  • Increment MINOR when adding an optional field or otherwise making a backward-compatible schema change.
  • Do not change the schema version for implementation fixes that leave the JSON contract unchanged.

Consumers should reject unsupported major versions and tolerate unknown fields within a supported major version.

Version History

1.0 — 2026-07-14

Added to root:

  • schema_version: str

Added to root.config:

  • test_id: str
  • test_name: str
  • description: str
  • test_scenario_name: str
  • system_path: str
  • tests_dir_path: str
  • test_scenario_path: str
  • container_image: str
  • cloudai_execution_node: str
  • env_vars: dict[str, Any]
  • gpus_per_node: Optional[int]
  • nodes: list[str]
  • clique_size: Optional[int]
  • fp8: Optional[str]
  • fp8_recipe: Optional[str]
  • expert_tensor_parallel_size: int

Training Report Models

TrainingResults

Container for parsed training output.

FieldTypeDefaultDescription
schema_versionstr'1.0'
configcloudai.report_generator.training.models.TrainingConfigRequired
stepsList[cloudai.report_generator.training.models.TrainingStep]Required
aggregationcloudai.report_generator.training.models.StepAggregation | NoneNone

TrainingConfig

Resolved training configuration from the framework artifact + CloudAI.

CloudAI-computed fields are supplied by the parser during construction.

FieldTypeDefaultDescription
test_idstrRequired
test_namestrRequired
descriptionstrRequired
test_scenario_namestrRequired
test_template_namestrRequired
system_pathstrRequired
tests_dir_pathstrRequired
test_scenario_pathstrRequired
container_imagestr''
cloudai_execution_nodestrRequired
env_varsdict[str, typing.Any]Required
world_sizeint | NoneNone
num_nodesintRequired
gpus_per_nodeint | NoneNone
nodeslist[str]Required
clique_sizeint | NoneNone
fp8str | NoneNone
fp8_recipestr | NoneNone
micro_batch_sizeintRequired
global_batch_sizeintRequired
seq_lengthintRequired
tensor_parallel_sizeintRequired
pipeline_parallel_sizeintRequired
context_parallel_sizeint | NoneRequired
virtual_pipeline_parallel_sizeint | NoneRequired
sequence_parallelboolRequired
expert_parallel_sizeintRequired
expert_tensor_parallel_sizeintRequired
data_parallel_sizeint | NoneNone
num_layersintRequired
hidden_sizeintRequired
num_attention_headsintRequired
num_query_groupsintRequired
ffn_hidden_sizeintRequired
kv_channelsintRequired
normalizationstrRequired
position_embedding_typestrRequired
model_namestr''
num_expertsint | NoneRequired
moe_router_topkint | NoneRequired
moe_ffn_hidden_sizeint | NoneRequired
moe_grouped_gemmbool | NoneRequired
profiling_enabledboolFalse
profiling_start_stepint | NoneNone
profiling_stop_stepint | NoneNone
exclude_start_stepsint5
exclude_post_profiling_stepsint2

TrainingStep

Results for a single training iteration.

FieldTypeDefaultDescription
iterationintRequired
step_time_secfloatRequired
lossfloatRequired
memory_reserved_bytesfloatRequired
memory_allocated_bytesfloatRequired
tflops_per_gpufloat | NoneNone

StepAggregation

Per-metric aggregated statistics over the filtered steps.

FieldTypeDefaultDescription
step_time_seccloudai.report_generator.training.models.MetricStatsRequired
losscloudai.report_generator.training.models.MetricStatsRequired
memory_reserved_bytescloudai.report_generator.training.models.MetricStatsRequired
memory_allocated_bytescloudai.report_generator.training.models.MetricStatsRequired
tflops_per_gpucloudai.report_generator.training.models.MetricStats | NoneNone

MetricStats

Aggregated statistics for one metric over the filtered steps.

FieldTypeDefaultDescription
meanfloatRequired
minfloatRequired
maxfloatRequired
stdfloatRequired
t99floatRequired
t95floatRequired