[Bugfix][Docs] Fix offline Whisper (vllm-project#13274)

Author: NickLucche

docs/source/models/supported_models.md

@@ -938,6 +938,26 @@ The following table lists those that are tested in vLLM.
   * ✅︎
 :::
 
+#### Transcription (`--task transcription`)
+
+Speech2Text models trained specifically for Automatic Speech Recognition.
+
+:::{list-table}
+:widths: 25 25 25 5 5
+:header-rows: 1
+
+- * Architecture
+  * Models
+  * Example HF Models
+  * [LoRA](#lora-adapter)
+  * [PP](#distributed-serving)
+- * `Whisper`
+  * Whisper-based
+  * `openai/whisper-large-v3-turbo`
+  * 🚧
+  * 🚧
+:::
+
 _________________
 
 ## Model Support Policy

vllm/entrypoints/llm.py

@@ -421,7 +421,7 @@ def generate(
             instead pass them via the ``inputs`` parameter.
         """
         runner_type = self.llm_engine.model_config.runner_type
-        if runner_type != "generate":
+        if runner_type not in ["generate", "transcription"]:
             messages = [
                 "LLM.generate() is only supported for (conditional) generation "
                 "models (XForCausalLM, XForConditionalGeneration).",