A high-throughput and memory-efficient inference and serving engine for LLMs
Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git
Synced Nov 8, 20:15 UTC
[Docs] Add bart-plugin to docs (#33905)
Signed-off-by: NickLucche <nlucches@redhat.com>
Nicolò Lucchesi committed
Feb 5, 2026 at 13:20 UTC
81a90e52776503c6cbdccd30fbe53f61c9179bdf
3 files changed
+17
-4
docs/design/plugin_system.md
+1
-1
@@ -45,7 +45,7 @@ Every plugin has three parts:
45
46
## Types of supported plugins
47
48
-- **General plugins** (with group name `vllm.general_plugins`): The primary use case for these plugins is to register custom, out-of-the-tree models into vLLM. This is done by calling `ModelRegistry.register_model` to register the model inside the plugin function.
48
+- **General plugins** (with group name `vllm.general_plugins`): The primary use case for these plugins is to register custom, out-of-the-tree models into vLLM. This is done by calling `ModelRegistry.register_model` to register the model inside the plugin function. For an example of an official model plugin, see the [bart-plugin](https://github.com/vllm-project/bart-plugin) which adds support for `BartForConditionalGeneration`.
49
50
- **Platform plugins** (with group name `vllm.platform_plugins`): The primary use case for these plugins is to register custom, out-of-the-tree platforms into vLLM. The plugin function should return `None` when the platform is not supported in the current environment, or the platform class's fully qualified name when the platform is supported.
51
docs/models/supported_models.md
+10
@@ -174,6 +174,16 @@ class MyConfig(PretrainedConfig):
174
- The `list` in the first element of the `tuple` contains the names of the input arguments
175
- The `list` in the last element of the `tuple` contains the names of the variables the layer outputs to in your modeling code
176
177
+### Plugins
178
+
179
+Some model architectures are supported via vLLM plugins. These plugins extend vLLM's capabilities through the [plugin system](../design/plugin_system.md).
180
+
181
+| Architecture | Models | Plugin Repository |
182
+|--------------|--------|-------------------|
183
+| `BartForConditionalGeneration` | BART | [bart-plugin](https://github.com/vllm-project/bart-plugin) |
184
+
185
+For other model architectures not natively supported, in particular for Encoder-Decoder models, we recommend following a similar pattern by implementing support through the plugin system.
186
+
187
## Loading a Model
188
189
### Hugging Face Hub
docs/usage/v1_guide.md
+6
-3
@@ -134,9 +134,12 @@ Please note that prefix caching is not yet supported for any of the above models
134
135
#### Encoder-Decoder Models
136
137
-Whisper is supported. Other models requiring cross-attention between separate
138
-encoder and decoder (e.g., `BartForConditionalGeneration`,
139
-`MllamaForConditionalGeneration`) are no longer supported.
137
+Whisper is supported natively. Other encoder-decoder models are supported via the plugin system:
138
+
139
+- **BART**: `BartForConditionalGeneration` is supported via the official [bart-plugin](https://github.com/vllm-project/bart-plugin).
140
+
141
+For other encoder-decoder models (e.g., `MllamaForConditionalGeneration`), we recommend
142
+following a similar pattern by implementing support through the [plugin system](../design/plugin_system.md).
143
144
### Features
145