EngineX-Hygon/sglang
5
0
Fork 0
Code Issues Pull Requests Actions 7 Projects Releases Wiki Activity

[FEAT] Add transformers backend support (#5929)

Browse Source
This commit is contained in:
Marc Sun
2025-06-04 06:05:29 +02:00
committed by GitHub
parent 8a5480528d
commit 37f1547587
11 changed files with 636 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files

1
docs/backend/server_arguments.md
View File

@@ -63,6 +63,7 @@ Please consult the documentation below and [server_args.py](https://github.com/s
| `kv_cache_dtype` | Dtype of the kv cache. | `auto` |
| `context_length` | The model's maximum context length. Defaults to None (will use the value from the model's config.json instead). Note that extending the default might lead to strange behavior. | None |
| `device` | The device we put the model. | None |
| `impl` | The implementation of the model to use. Defaults to SGlang implementation and fall back to transformers if needed | `auto` |
| `served_model_name` | Override the model name returned by the v1/models endpoint in OpenAI API server.| None |
| `is_embedding` | Set to `true` to perform [embedding](./openai_api_embeddings.ipynb) / [encode](https://docs.sglang.ai/backend/native_api#Encode-(embedding-model)) and [reward](https://docs.sglang.ai/backend/native_api#Classify-(reward-model)) tasks. | `False` |
| `revision` | Adjust if a specific version of the model should be used. | None |