[BugFix]Add MiniCPM-o 4.5 end-to-end test suite (offline, online serving) (#5237)
Signed-off-by: y_null <y_null@qq.com> Signed-off-by: amy-why-3459 <wuhaiyan17@huawei.com> Co-authored-by: amy-why-3459 <wuhaiyan17@huawei.com> Co-authored-by: Hongsheng Liu <liuhongsheng4@huawei.com>
This commit is contained in:
27 files changed
+1338
-46
No files matched your search
@@ -6,7 +6,7 @@ Two-stage pipeline: **thinker** (multimodal understanding) → **talker + Token2
|
||||
|
||||
## Setup
|
||||
|
||||
- Install talker deps: `pip install 'vllm-omni[minicpmo]'` (or `stepaudio2-minicpmo`)
|
||||
- Install talker deps: `pip install stepaudio2-minicpmo`
|
||||
- `--trust-remote-code` is always passed by `end2end.py` (required for MiniCPMO)
|
||||
- See [stage configuration docs](https://docs.vllm.ai/projects/vllm-omni/en/latest/configuration/stage_configs/) for memory tuning
|
||||
|
||||
|
||||
@@ -13,10 +13,7 @@ capability boundary, and validation scope, see
|
||||
Install vLLM-Omni with the MiniCPM-o talker dependencies:
|
||||
|
||||
```bash
|
||||
pip install 'vllm-omni[minicpmo]'
|
||||
|
||||
# From a source checkout:
|
||||
pip install -e '.[minicpmo]'
|
||||
pip install stepaudio2-minicpmo
|
||||
```
|
||||
|
||||
The `minicpmo` extra installs `stepaudio2-minicpmo` and its audio dependencies,
|
||||
|
||||
Reference in new issue
Block a user