[doc] feat: add rollout&train consistency doc for Ascend Platform (#4166)

### What does this PR do?
This document provides guidance on ensuring consistency between verl and
vllm inference results on NPU.

### 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`, `sglang`, `vllm`, `rollout`,
`trainer`, `ci`, `training_utils`, `recipe`, `hardware`, `deployment`,
`ray`, `worker`, `single_controller`, `misc`, `perf`, `model`, `algo`,
`env`, `tool`, `ckpt`, `doc`, `data`
- 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.

```python
# 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](https://github.com/volcengine/verl/blob/main/CONTRIBUTING.md).
- [ ] Apply [pre-commit
checks](https://github.com/volcengine/verl/blob/main/CONTRIBUTING.md#code-linting-and-formatting):
`pre-commit install && pre-commit run --all-files --show-diff-on-failure
--color=always`
- [ ] Add / Update [the
documentation](https://github.com/volcengine/verl/tree/main/docs).
- [ ] Add unit or end-to-end test(s) to [the CI
workflow](https://github.com/volcengine/verl/tree/main/.github/workflows)
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](https://verl-project.slack.com/archives/C091TCESWB1) in [the
`verl` Slack
workspace](https://join.slack.com/t/verl-project/shared_invite/zt-3855yhg8g-CTkqXu~hKojPCmo7k_yXTQ).
(If not accessible, please try [the Feishu group
(飞书群)](https://applink.larkoffice.com/client/chat/chatter/add_by_link?link_token=772jd4f1-cd91-441e-a820-498c6614126a).)

Co-authored-by: wangxiaoxin-sherie <[email protected]>

Author: momo609

docs/ascend_tutorial/ascend_consistency.rst

@@ -0,0 +1,50 @@
+Align the Inference results of the verl and vLLM frameworks on Ascend devices(zh)
+====================================
+
+在昇腾设备上对齐verl和vLLM两个框架下的推理结果。
+
+Last updated: 11/17/2025.
+
+这是一份在昇腾设备上对齐verl和vLLM两个框架下推理结果的教程。
+
+环境变量配置
+~~~~~~~~~~~~
+
+在多卡通信情况下：
+
+- HCCL通信下(默认场景):
+  
+  -  export CLOSE_MATMUL_K_SHIFT=1
+  -  export ATB_MATMUL_SHUFFLE_K_ENABLE=0
+  -  export HCCL_DETERMINISTIC="true"
+  -  export VLLM_ENABLE_V1_MULTIPROCESSING=0
+
+- LCCL通信下(通过export HCCL_OP_EXPANSION_MODE="AIV"使能）:
+
+  -  export CLOSE_MATMUL_K_SHIFT=1
+  -  export ATB_MATMUL_SHUFFLE_K_ENABLE=0
+  -  export LCCL_DETERMINISTIC=1
+  -  export ATB_LLM_LCOC_ENABLE=0
+  -  export VLLM_ENABLE_V1_MULTIPROCESSING=0
+
+在单卡无通信情况下：
+
+- HCCL和LCCL通信下:
+ 
+  -  export CLOSE_MATMUL_K_SHIFT=1
+  -  export ATB_MATMUL_SHUFFLE_K_ENABLE=0
+  -  export VLLM_ENABLE_V1_MULTIPROCESSING=0
+
+vLLM初始化参数
+~~~~~~~~~~~~
+
+需要对 SamplingParams 参数里单独设置seed, 保持vLLM和verl推理结果一致, 举例修改如下：
+
+.. code:: yaml
+
+      sampling_params = SamplingParams(n=1,
+                                       logprobs=0,  # can be set to 0 and let actor to recompute
+                                       max_tokens=config.response_length,
+                                       repetition_penalty=config.get("repetition_penalty", 1.0),
+                                       seed=1234)
+

docs/index.rst

@@ -140,6 +140,7 @@ verl is fast with:
    amd_tutorial/amd_build_dockerfile_page.rst
    amd_tutorial/amd_vllm_page.rst
    ascend_tutorial/ascend_quick_start.rst
+   ascend_tutorial/ascend_consistency.rst
    ascend_tutorial/ascend_profiling_zh.rst
    ascend_tutorial/ascend_profiling_en.rst
    ascend_tutorial/dockerfile_build_guidance.rst