EngineX/xc-llm-ascend
3
0
Fork 0
Code Issues Pull Requests Actions Projects Releases Wiki Activity

[V1][Core] Add support for V1 Engine (#295)

Browse Source 
### What this PR does / why we need it?
Add support for V1 Engine.

Please note that this is just the initial version, and there may be some
places need to be fixed or optimized in the future, feel free to leave
some comments to us.

### Does this PR introduce _any_ user-facing change?

To use V1 Engine on NPU device, you need to set the env variable shown
below:

```bash
export VLLM_USE_V1=1
export VLLM_WORKER_MULTIPROC_METHOD=spawn
```

If you are using vllm for offline inferencing, you must add a `__main__`
guard like:

```bash
if __name__ == '__main__':

    llm = vllm.LLM(...)
```

Find more details
[here](https://docs.vllm.ai/en/latest/getting_started/troubleshooting.html#python-multiprocessing).

### How was this patch tested?
I have tested the online serving with `Qwen2.5-7B-Instruct` using this
command:

```bash
vllm serve Qwen/Qwen2.5-7B-Instruct --max_model_len 26240
```

Query the model with input prompts:

```bash
curl http://localhost:8000/v1/completions \
    -H "Content-Type: application/json" \
    -d '{
        "model": "Qwen/Qwen2.5-7B-Instruct",
        "prompt": "The future of AI is",
        "max_tokens": 7,
        "temperature": 0
    }'
```

---------

Signed-off-by: shen-shanshan <467638484@qq.com>
Co-authored-by: didongli182 <didongli@huawei.com>
This commit is contained in:
Shanshan Shen
2025-03-20 19:34:44 +08:00
committed by GitHub
parent 663dca7578
commit c06af8b2e0
13 changed files with 1385 additions and 38 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
docs/source/installation.md
View File

@@ -64,7 +64,7 @@ docker run --rm \
-it $IMAGE bash
```
:::{dropdown} Click here to see "Install CANN manally"
:::{dropdown} Click here to see "Install CANN manually"
:animate: fade-in-slide-down
You can also install CANN manually:

2
docs/source/tutorials/multi_node.md
View File

@@ -56,7 +56,7 @@ ray start --address='{head_node_ip}:{port_num}' --num-gpus=8 --node-ip-address={
```
:::{note}
If you're running DeepSeek V3/R1, please remove `quantization_config` section in `config.json` file since it's not supported by vllm-ascend currentlly.
If you're running DeepSeek V3/R1, please remove `quantization_config` section in `config.json` file since it's not supported by vllm-ascend currently.
:::
Start the vLLM server on head node:

10
docs/source/user_guide/release_notes.md
View File

@@ -25,8 +25,8 @@
- Pin modelscope<1.23.0 on vLLM v0.7.3 to resolve: https://github.com/vllm-project/vllm/pull/13807
### Known issues
- In [some cases](https://github.com/vllm-project/vllm-ascend/issues/324), expecially when the input/output is very long, the accuracy of output may be incorrect. We are working on it. It'll be fixed in the next release.
- Improved and reduced the garbled code in model output. But if you still hit the issue, try to change the gerneration config value, such as `temperature`, and try again. There is also a knonwn issue shown below. Any [feedback](https://github.com/vllm-project/vllm-ascend/issues/267) is welcome. [#277](https://github.com/vllm-project/vllm-ascend/pull/277)
- In [some cases](https://github.com/vllm-project/vllm-ascend/issues/324), especially when the input/output is very long, the accuracy of output may be incorrect. We are working on it. It'll be fixed in the next release.
- Improved and reduced the garbled code in model output. But if you still hit the issue, try to change the generation config value, such as `temperature`, and try again. There is also a knonwn issue shown below. Any [feedback](https://github.com/vllm-project/vllm-ascend/issues/267) is welcome. [#277](https://github.com/vllm-project/vllm-ascend/pull/277)
## v0.7.1rc1
@@ -46,7 +46,7 @@ Please follow the [official doc](https://vllm-ascend.readthedocs.io/en/v0.7.1-de
### Core
- Added the Ascend quantization config option, the implementation will comming soon. [#7](https://github.com/vllm-project/vllm-ascend/pull/7) [#73](https://github.com/vllm-project/vllm-ascend/pull/73)
- Added the Ascend quantization config option, the implementation will coming soon. [#7](https://github.com/vllm-project/vllm-ascend/pull/7) [#73](https://github.com/vllm-project/vllm-ascend/pull/73)
- Add silu_and_mul and rope ops and add mix ops into attention layer. [#18](https://github.com/vllm-project/vllm-ascend/pull/18)
### Other
@@ -58,5 +58,5 @@ Please follow the [official doc](https://vllm-ascend.readthedocs.io/en/v0.7.1-de
### Known issues
- This release relies on an unreleased torch_npu version. It has been installed within official container image already. Please [install](https://vllm-ascend.readthedocs.io/en/v0.7.1rc1/installation.html) it manually if you are using non-container environment.
- There are logs like `No platform deteced, vLLM is running on UnspecifiedPlatform` or `Failed to import from vllm._C with ModuleNotFoundError("No module named 'vllm._C'")` shown when runing vllm-ascend. It actually doesn't affect any functionality and performance. You can just ignore it. And it has been fixed in this [PR](https://github.com/vllm-project/vllm/pull/12432) which will be included in v0.7.3 soon.
- There are logs like `# CPU blocks: 35064, # CPU blocks: 2730` shown when runing vllm-ascend which should be `# NPU blocks:` . It actually doesn't affect any functionality and performance. You can just ignore it. And it has been fixed in this [PR](https://github.com/vllm-project/vllm/pull/13378) which will be included in v0.7.3 soon.
- There are logs like `No platform detected, vLLM is running on UnspecifiedPlatform` or `Failed to import from vllm._C with ModuleNotFoundError("No module named 'vllm._C'")` shown when running vllm-ascend. It actually doesn't affect any functionality and performance. You can just ignore it. And it has been fixed in this [PR](https://github.com/vllm-project/vllm/pull/12432) which will be included in v0.7.3 soon.
- There are logs like `# CPU blocks: 35064, # CPU blocks: 2730` shown when running vllm-ascend which should be `# NPU blocks:` . It actually doesn't affect any functionality and performance. You can just ignore it. And it has been fixed in this [PR](https://github.com/vllm-project/vllm/pull/13378) which will be included in v0.7.3 soon.

2
docs/source/user_guide/suppoted_features.md
View File

@@ -9,7 +9,7 @@
| Speculative decoding | ✅ | | | Basic functions available | Need fully test |
| Pooling | ✅ | | | Basic functions available(Bert) | Need fully test and add more models support|
| Enc-dec | ❌ | | | NA | Plan in 2025.06.30|
| Multi Modality | ✅ | | ✅ | Basic functions available(LLaVA/Qwen2-vl/Qwen2-audio/internVL)| Improve perforamance, and add more models support |
| Multi Modality | ✅ | | ✅ | Basic functions available(LLaVA/Qwen2-vl/Qwen2-audio/internVL)| Improve performance, and add more models support |
| LogProbs | ✅ | | | Basic functions available | Need fully test |
| Prompt logProbs | ✅ | | | Basic functions available | Need fully test |
| Async output | ✅ | | | Basic functions available | Need fully test |