v0.7.3 - 2026-10-16

This page is for version Latest (default). For other versions, use one of these documentation indexes: - Latest (default): https://docs.nvidia.com/nemo-helix/latest/llms.txt - 0.8.0: https://docs.nvidia.com/nemo-helix/v0.8.0/llms.txt - 0.7.3: https://docs.nvidia.com/nemo-helix/v0.7.3/llms.txt - 0.7.2: https://docs.nvidia.com/nemo-helix/v0.7.2/llms.txt - 0.7.1: https://docs.nvidia.com/nemo-helix/v0.7.1/llms.txt - 0.7.0: https://docs.nvidia.com/nemo-helix/v0.7.0/llms.txt - 0.6.0: https://docs.nvidia.com/nemo-helix/v0.6.0/llms.txt - 0.5.1: https://docs.nvidia.com/nemo-helix/v0.5.1/llms.txt - 0.5.0: https://docs.nvidia.com/nemo-helix/v0.5.0/llms.txt - 0.4.0: https://docs.nvidia.com/nemo-helix/v0.4.0/llms.txt - 0.3.0: https://docs.nvidia.com/nemo-helix/v0.3.0/llms.txt - 0.2.0: https://docs.nvidia.com/nemo-helix/v0.2.0/llms.txt

NeMo Helix v0.7.3 is a patch release for the 0.7 release line that helps existing Helm installations preserve embedded PostgreSQL storage metadata during upgrades. It also includes the deployment readiness fixes in v0.7.2.

What’s fixed

• Preserve PostgreSQL claim-template labels. The Helm chart adds postgresql.persistence.claimTemplateLabels so older installations can retain the exact labels on the embedded PostgreSQL StatefulSet’s data volume claim template. This avoids upgrade failures caused by changes to immutable claim-template labels. The default {} preserves existing chart behavior; operators must supply their existing labels when needed.

Upgrade from v0.7.2

If you use embedded PostgreSQL, copy the complete label map from the existing StatefulSet’s data claim template into postgresql.persistence.claimTemplateLabels before upgrading. Keep those exact values on subsequent upgrades, including old chart and application version labels. This setting only addresses claim-template labels; preserve existing release and resource names, storage configuration, and the secrets encryption key and its Secret reference as well. See Database Setup.

Known issues

• DPO jobs fail before the first training step with KeyError: 'checkpoint'. The generated DPO configuration still omits the policy.dtensor_cfg.checkpoint section required by the pinned NeMo-RL version. There is no workaround in the 0.7 release line. DPO jobs run on v0.6.0; GRPO jobs are not affected. See Customizer troubleshooting.
• Hermes trials can omit the relay_atif evidence descriptor. Use the trace:atif descriptor instead; it points at the same ATIF file.