minke.yu/sglang
1
0
Fork 0
Code Issues Pull Requests Actions 12 Packages Projects Releases Wiki Activity

Add pr tests (#31952)

Browse Source 
Co-authored-by: Claude <noreply@anthropic.com>
Co-authored-by: sglang-npu-bot <sglangnpu@163.com>
Co-authored-by: Cherry_ming <136634645@qq.com>
This commit is contained in:
Sugar920
2026-08-01 15:03:21 +08:00
committed by GitHub
co-authored by Claude sglang-npu-bot Cherry_ming
parent fd96a35fb0
commit 33ecf4bcd8
170 changed files with 1968 additions and 329 deletions
Download Patch File Download Diff File Expand all files Collapse all files
docs_new/docs/hardware-platforms/ascend-npus/development/contribution_guide.mdx
+4 -4
View File
@@ -42,7 +42,7 @@ pre-commit run --all-files
## Run and add tests
All NPU tests are end-to-end (E2E) and require launching a server with real model weights.
Tests live under [`test/registered/ascend/`](https://github.com/sgl-project/sglang/tree/main/test/registered/ascend), organized by model type and functionality:
Tests live under [`test/registered/npu/`](https://github.com/sgl-project/sglang/tree/main/test/registered/npu), organized by model type and functionality:
```
ascend/
@@ -58,9 +58,9 @@ ascend/
### Adding a test
See [`test_npu_sampling_backend.py`](https://github.com/sgl-project/sglang/tree/main/test/registered/ascend/basic_function/backends/test_npu_sampling_backend.py) for a complete example. Key steps:
See [`test_npu_sampling_backend.py`](https://github.com/sgl-project/sglang/tree/main/test/registered/npu/basic_function/backends/test_npu_sampling_backend.py) for a complete example. Key steps:
1. Place your test file in the appropriate directory under `test/registered/ascend/`.
1. Place your test file in the appropriate directory under `test/registered/npu/`.
2. Extend `CustomTestCase` (from `sglang.test.test_utils`) for CI retry support.
3. Launch server with `popen_launch_server()` in `setUpClass` and clean up with `kill_process_tree()` in `tearDownClass`.
4. Register your test with `register_npu_ci()`:
@@ -73,7 +73,7 @@ See [`test_npu_sampling_backend.py`](https://github.com/sgl-project/sglang/tree/
### Running tests locally
```bash
pytest test/registered/ascend/llm_models/test_npu_qwen3_0_6b.py -v
pytest test/registered/npu/llm_models/test_npu_qwen3_0_6b.py -v
```
For detailed instructions, refer to [test/README.md](https://github.com/sgl-project/sglang/tree/main/test/README.md).
docs_new/docs/hardware-platforms/ascend-npus/development/support_new_models.mdx
+1 -1
View File
@@ -71,7 +71,7 @@ Please note all your testing and benchmarking results in PR description.
- **(Optional) Other evals**: If you ran other evals, please note the results in PR description.
<Tip>
For NPU-adapted models: add the corresponding test under `test/registered/ascend/` and verify correctness on Ascend NPU
For NPU-adapted models: add the corresponding test under `test/registered/npu/` and verify correctness on Ascend NPU
hardware; run benchmarks on the NPU device and report performance metrics (TTFT, throughput), comparing against SGLang
GPU results as the primary baseline. Fall back to HF Transformer comparison when no GPU adaptation is available.
</Tip>