CloudAI Benchmark Framework v1.5.0

Workloads Documentation

This section contains automatically generated documentation for all CloudAI workloads. Each workload provides specific functionality for running different types of tests and benchmarks.

Test

Slurm

Kubernetes

RunAI

Standalone

AIConfigurator ❌ ❌ ❌ ✅
AI Dynamo ✅ ✅ ❌ ❌
Bash Command ✅ ❌ ❌ ❌
Chakra Replay ✅ ❌ ❌ ❌
DDLB ✅ ❌ ❌ ❌
DeepEP Benchmark ✅ ❌ ❌ ❌
JaxToolbox workloads (DEPRECATED) ✅ ❌ ❌ ❌
MegatronRun ✅ ❌ ❌ ❌
MegatronBridge ✅ ❌ ❌ ❌
NCCL ✅ ✅ ✅ ❌
NeMo v1.0 aka NemoLauncher (DEPRECATED) ✅ ❌ ❌ ❌
Nemo Run ✅ ❌ ❌ ❌
NIXL Bench ✅ ❌ ❌ ❌
NIXL KVBench ✅ ❌ ❌ ❌
NIXL CTPerf ✅ ❌ ❌ ❌
Sleep ✅ ✅ ❌ ✅
Slurm Container ✅ ❌ ❌ ❌
Triton Inference ✅ ❌ ❌ ❌
UCC ✅ ❌ ❌ ❌

To add documentation for a new workload:

  1. Add docstrings to your Python classes and methods.

  2. Create a reStructuredText file in doc/workloads/ (e.g., my_workload.rst).

  3. Add it to the table above.

The documentation will be automatically generated during the build process.

Previous Development
Next AI Dynamo
© Copyright 2026, NVIDIA CORPORATION & AFFILIATES. Last updated on Mar 3, 2026