Streaming tool call deltas support

[nvcsathe]

• I, the PR author, have personally reviewed every line of this PR.

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 @NVIDIA/mcore-oncall.

Issue tracking

For PRs from open-source community contributors:

• New features: a linked issue is required. Please open a feature request and reference it here before submitting the PR.
• Small updates (bug fixes, minor improvements): a linked issue is recommended and will accelerate the PR review process.

Linked issue:

Contribution process

Pre-checks

• 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

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

All PRs start as draft. If you open a non-draft PR, it will be automatically converted to draft.

Step 1: Mark PR as "Ready for Review"

1. When your PR is ready, click Ready for Review.
2. An oncall reviewer is auto-assigned and expert reviewers are notified based on your changes.
   • Some PRs may jump straight to step 2. This is determined by .github/CODEOWNERS.

⚠️ Only mark as ready once merge-conflicts are resolved and the CI is passing.
Final Review might get declined if these requirements are not fulfilled.

Step 2: Final Review

For PRs that change megatron/core, once all expert reviewers have approved, the Final Review label is applied automatically and final reviewers are assigned.

For PRs outside megatron/core, this step is skipped.

Step 3: Approved

Once all required reviewers have approved, the Approved label is applied automatically.

Merge

Any member of mcore-engineers will be able to merge your PR.

[copy-pr-bot]

Auto-sync is disabled for draft pull requests in this repository. Workflows must be run manually.

Contributors can view more details about this message here.

[shanmugamr1992]

NeMo-RL functional tests

mcore sha: 0b57f670b94d7fdc0a0f77b0d0301782086a3578 · baseline main sha: 19c50e864b101abbe7f4e57262ddf5b6daff055d · bridge sha: 2940a635731da68c5960a020f8ea2f678fa1e073 · nemo rl sha: 86774472e229b457314418fe3f25e8e844775596 · suite: L1_Functional_Tests_Megatron_4 · cluster: oci-hsg · updated 2026-08-06 05:26 UTC

Test	Status	Comment
grpo_megatron_generation	🛠️ fixed	Passes only with the Megatron-Bridge fix in NVIDIA-NeMo/Megatron-Bridge#5357 applied. Without it this test fails at import time on megatron-core main too, so the break is not caused by this PR.
grpo_megatron_generation_non_colocated	🛠️ fixed	Passes only with the Megatron-Bridge fix in NVIDIA-NeMo/Megatron-Bridge#5357 applied. Without it this test fails at import time on megatron-core main too, so the break is not caused by this PR.
grpo_megatron_generation_async	🛠️ fixed	Passes only with the Megatron-Bridge fix in NVIDIA-NeMo/Megatron-Bridge#5357 applied. Without it this test fails at import time on megatron-core main too, so the break is not caused by this PR.
grpo_megatron_generation_colocated_async	🛠️ fixed	Passes only with the Megatron-Bridge fix in NVIDIA-NeMo/Megatron-Bridge#5357 applied. Without it this test fails at import time on megatron-core main too, so the break is not caused by this PR.
grpo_megatron_generation_async_gym	🛠️ fixed	Passes only with the Megatron-Bridge fix in NVIDIA-NeMo/Megatron-Bridge#5357 applied. Without it this test fails at import time on megatron-core main too, so the break is not caused by this PR.
grpo_megatron_generation_topp_topk	⚠️ fail (pre-existing)	Pre-existing on megatron-core main, not caused by any labeled PR. grpo_megatron_generation_topp_topk runs to completion but max(train/token_mult_prob_error) lands at 1.078047513961792 against a < 1.06 threshold. The value is bit-identical on Megatron-LM#5382 head eb01b68 and on main 3aee84c (both with the Bridge FSDP fix), so it is a deterministic train/generation logprob mismatch in the top-p/top-k sampling path rather than run-to-run noise. Marginal at ~1.7% over. No fix raised; needs an owner to decide whether the sampling path regressed or the threshold is stale.
grpo_megatron_generation_colocated_gym	🛠️ fixed	Passes only with the Megatron-Bridge fix in NVIDIA-NeMo/Megatron-Bridge#5357 applied. Without it this test fails at import time on megatron-core main too, so the break is not caused by this PR.

No failure in this suite is attributable to this PR.

L1 is currently blocked for every PR by a Megatron-Bridge break: Megatron-LM#5865 turned mcore_fsdp_adapter.FullyShardedDataParallel into a factory function, so Bridge's isinstance() checks against it raise TypeError. The fix is in draft at NVIDIA-NeMo/Megatron-Bridge#5357 and this run carried it — that is why six rows read 🛠️ fixed rather than plain pass. Until it merges these tests fail at import on main too, so nothing here is a signal about your change.

grpo_megatron_generation_topp_topk fails identically on a megatron-core main baseline (19c50e86, same Bridge and NeMo-RL), missing a metric threshold rather than crashing. Pre-existing and tracked in NVIDIA-NeMo/RL#3365.

L2 has not been run yet.

_{Posted by the nemo-rl-testing-agent. Re-runs edit this comment in place.}

[svcnvidia-nemo-ci]

🔄 Merge queue validation started!

You can track the progress here: https://github.com/NVIDIA/Megatron-LM/actions/runs/31752201591

[svcnvidia-nemo-ci]

🔄 Merge queue validation started!

You can track the progress here: https://github.com/NVIDIA/Megatron-LM/actions/runs/31752551935

[svcnvidia-nemo-ci]

🔄 Merge queue validation started!

You can track the progress here: https://github.com/NVIDIA/Megatron-LM/actions/runs/31755641943

[svcnvidia-nemo-ci]

🔄 Merge queue validation started!

You can track the progress here: https://github.com/NVIDIA/Megatron-LM/actions/runs/31764524168