[Misc] Reorganize inputs (#35182)

Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>

docs/api/README.md

@@ -27,11 +27,9 @@ LLM Class.
 
 - [vllm.LLM][]
 
-LLM Inputs.
+Prompt schema for LLM APIs.
 
-- [vllm.inputs.PromptType][]
-- [vllm.inputs.TextPrompt][]
-- [vllm.inputs.TokensPrompt][]
+- [vllm.inputs.llm][]
 
 ## vLLM Engines
 
@@ -58,13 +56,7 @@ Looking to add your own multi-modal model? Please follow the instructions listed
 
 - [vllm.multimodal.MULTIMODAL_REGISTRY][]
 
-### Inputs
-
-User-facing inputs.
-
-- [vllm.multimodal.inputs.MultiModalDataDict][]
-
-Internal data structures.
+### Internal data structures
 
 - [vllm.multimodal.inputs.PlaceholderRange][]
 - [vllm.multimodal.inputs.NestedTensors][]
@@ -72,7 +64,6 @@ Internal data structures.
 - [vllm.multimodal.inputs.MultiModalFieldConfig][]
 - [vllm.multimodal.inputs.MultiModalKwargsItem][]
 - [vllm.multimodal.inputs.MultiModalKwargsItems][]
-- [vllm.multimodal.inputs.MultiModalInputs][]
 
 ### Data Parsing
 

docs/contributing/model/transcription.md

@@ -23,7 +23,7 @@ Declare supported languages and capabilities:
     from torch import nn
 
     from vllm.config import ModelConfig, SpeechToTextConfig
-    from vllm.inputs.data import PromptType
+    from vllm.inputs import PromptType
     from vllm.model_executor.models.interfaces import SupportsTranscription
     
     class YourASRModel(nn.Module, SupportsTranscription):
@@ -66,7 +66,7 @@ This is for controlling general behavior of the API when serving your model:
 
 See [Audio preprocessing and chunking](#audio-preprocessing-and-chunking) for what each field controls.
 
-Implement the prompt construction via [get_generation_prompt][vllm.model_executor.models.interfaces.SupportsTranscription.get_generation_prompt]. The server passes you the resampled waveform and task parameters; you return a valid [PromptType][vllm.inputs.data.PromptType]. There are two common patterns:
+Implement the prompt construction via [get_generation_prompt][vllm.model_executor.models.interfaces.SupportsTranscription.get_generation_prompt]. The server passes you the resampled waveform and task parameters; you return a valid [PromptType][vllm.inputs.llm.PromptType]. There are two common patterns:
 
 #### Multimodal LLM with audio embeddings (e.g., Voxtral, Gemma3n)
 

docs/features/multimodal_inputs.md

@@ -18,7 +18,7 @@ This page teaches you how to pass multi-modal inputs to [multi-modal models](../
 To input multi-modal data, follow this schema in [vllm.inputs.PromptType][]:
 
 - `prompt`: The prompt should follow the format that is documented on HuggingFace.
-- `multi_modal_data`: This is a dictionary that follows the schema defined in [vllm.multimodal.inputs.MultiModalDataDict][].
+- `multi_modal_data`: This is a dictionary that follows the schema defined in [vllm.inputs.MultiModalDataDict][].
 
 ### Image Inputs