ci: Mount and enforce HF_HOME

[ko3n1g]

What does this PR do ?

⚠️ For major changes (either in lines of code or in its impact), please make sure to first share a design doc with the team. If you're unsure what's the best way to do so, contact the @mcore-oncall.

Contribution process

flowchart LR
    A[Pre-checks] --> B[PR Tests]
    subgraph Code Review/Approval
        C1[Expert Review] --> C2[Final Review]
    end
    B --> C1
    C2 --> D[Merge]

Loading

Pre-checks

• I want this PR in a versioned release and have added the appropriate Milestone (e.g., Core 0.8)
• I have added relevant unit tests
• I have added relevant functional tests
• I have added proper typing to my code Typing guidelines
• I have added relevant documentation
• I have run the autoformatter.sh on my PR

Code review

The following process is enforced via the CODEOWNERS file for changes into megatron/core. For changes outside of megatron/core, it is up to the PR author whether or not to tag the Final Reviewer team.

For MRs into `main` branch

Feel free to message or comment the @mcore-oncall to help accelerate your merge into main. The less complex your PR is, the faster it will be approved and merged!

(Step 1): Add PR label Expert Review

(Step 2): Collect the expert reviewers reviews

1. Attach the Expert Review label when your PR is ready for review.
2. GitHub auto-assigns expert reviewers based on your changes. They will get notified and pick up your PR soon.

⚠️ Only proceed to the next step once all reviewers have approved, merge-conflict are resolved and the CI is passing.
Final Review might get declined if these requirements are not fulfilled.

(Step 3): Final Review

1. Add Final Review label
2. GitHub auto-assigns final reviewers based on your changes. They will get notified and pick up your PR soon.

(Optional Step 4): Cherry-pick into release branch

If this PR also needs to be merged into core_r* release branches, after this PR has been merged, select Cherry-pick to open a new PR into the release branch.

For MRs into `dev` branch

The proposed review process for `dev` branch is under active discussion.

MRs are mergable after one approval by either eharper@nvidia.com or zijiey@nvidia.com.

Merging your PR

Any member of core-adlr and core-nemo will be able to merge your PR.

[ko3n1g]

/claude review

[ko3n1g]

/claude review

[claude]

Code Review

Bug: response shape change for previously-optional fields

The old code only included policy_staleness, kv_cache_staleness, and num_evictions in the response when they had meaningful values. The new code unconditionally includes them:

• policy_staleness / kv_cache_staleness will be None for any request where those tensors were never populated (they're Optional[torch.Tensor] = None on DynamicInferenceRequest).
• num_evictions will be 0 even when there were no eviction events.

This is a breaking change to the response schema — consumers that treat the absence of these keys as "not applicable" will now always see them. If always-present fields are intentional, that's fine, but it should be explicit and clients should be updated.

Potential TypeError from chat_template_kwargs overlap

prompt_tokens = tokenizer.apply_chat_template(
    messages,
    tokenize=True,
    add_generation_prompt=True,
    tools=req.get("tools", None),
    **req.get("chat_template_kwargs", {}),
)

If a caller passes tokenize, add_generation_prompt, or tools inside chat_template_kwargs, Python raises TypeError: got multiple values for keyword argument. There is no validation to guard against this — consider an explicit check or stripping overlapping keys with a warning.

Missing test coverage

No tests cover the chat_completions endpoint response shape for these changed fields. The existing tests in tests/unit_tests/inference/engines/test_dynamic_engine.py exercise DynamicInferenceRequest directly, not the endpoint. Missing coverage:

• policy_staleness / kv_cache_staleness always present (possibly None) in response
• num_evictions always 0 when there are no EVICT events
• chat_template_kwargs forwarding, including the overlap TypeError case

[svcnvidia-nemo-ci]

🔄 Merge queue validation started!

You can track the progress here: https://github.com/NVIDIA/Megatron-LM/actions/runs/22685751754

[svcnvidia-nemo-ci]

🔄 Merge queue validation started!

You can track the progress here: https://github.com/NVIDIA/Megatron-LM/actions/runs/22685854327

[svcnvidia-nemo-ci]

🔄 Merge queue validation started!

You can track the progress here: https://github.com/NVIDIA/Megatron-LM/actions/runs/22692993456

[svcnvidia-nemo-ci]

🔄 Merge queue validation started!

You can track the progress here: https://github.com/NVIDIA/Megatron-LM/actions/runs/22695852416