[model] feat: add qwen3-122b long seq launch script for ascend

[zjchenn]

What does this PR do?

Add the Ascend Megatron GRPO script for Qwen3.5-122B-A10B long-sequence training.

Checklist Before Starting

• Search for similar PRs. Paste at least one query link here: ...
• Format the PR title as [{modules}] {type}: {description} (This will be checked by the CI)
  • {modules} include fsdp, megatron, veomni, sglang, vllm, rollout, trainer, ci, training_utils, recipe, hardware, deployment, ray, worker, single_controller, misc, perf, model, algo, env, tool, ckpt, doc, data, cfg, reward, fully_async, one_step_off
  • If this PR involves multiple modules, separate them with , like [megatron, fsdp, doc]
  • {type} is in feat, fix, refactor, chore, test
  • If this PR breaks any API (CLI arguments, config, function signature, etc.), add [BREAKING] to the beginning of the title.
  • Example: [BREAKING][fsdp, megatron] feat: dynamic batching

Test

For changes that can not be tested by CI (e.g., algorithm implementation, new model support), validate by experiment(s) and show results like training curve plots, evaluation results, etc.

API and Usage Example

Demonstrate how the API changes if any, and provide usage example(s) if possible.

# Add code snippet or script demonstrating how to use this

Design & Code Changes

Demonstrate the high-level design if this PR is complex, and list the specific changes.

Checklist Before Submitting

Important

Please check all the following items before requesting a review, otherwise the reviewer might deprioritize this PR for review.

• Read the Contribute Guide.
• Apply pre-commit checks: pre-commit install && pre-commit run --all-files --show-diff-on-failure --color=always
• Add / Update the documentation.
• Add unit or end-to-end test(s) to the CI workflow to cover all the code. If not feasible, explain why: ...
• Once your PR is ready for CI, send a message in the ci-request channel in the verl Slack workspace. (If not accessible, please try the Feishu group (飞书群).)
• If your PR is related to the recipe submodule, please also update the reference to the submodule commit via git submodule update --remote or cd recipe && git pull origin main.

[gemini-code-assist]

Code Review

This pull request introduces a new launch script for running GRPO training with Qwen3.5-122B using Megatron and vLLM on Ascend NPUs. A critical issue was identified where the undefined variable NDEVICES_PER_NODE is referenced, which will cause the script to crash immediately due to the set -u option. It is recommended to replace it with the defined NPUS_PER_NODE variable.

Important

The consumer version of Gemini Code Assist on GitHub is being sunset. Starting June 18, 2026, new organization installations will be blocked, and all code review activity will officially cease on July 17, 2026.
For more details on the timeline and next steps, please review the Help Documentation.

[gemini-code-assist]

The variable NDEVICES_PER_NODE is not defined in this script. Since set -u is enabled, referencing an unbound variable will cause the script to crash immediately upon execution. It should be replaced with NPUS_PER_NODE which is defined on line 17.

Suggested change

	trainer.n_gpus_per_node=${NDEVICES_PER_NODE}
	trainer.n_gpus_per_node=${NPUS_PER_NODE}

References

1. Enable set -xeuo pipefail in shell scripts to ensure that the script exits on errors, treats unset variables as errors, and pipelines fail correctly.