[Docs] Document Deepseek V3 support (#11535)

Signed-off-by: simon-mo <[email protected]>

README.md

@@ -60,7 +60,7 @@ vLLM is flexible and easy to use with:
 
 vLLM seamlessly supports most popular open-source models on HuggingFace, including:
 - Transformer-like LLMs (e.g., Llama)
-- Mixture-of-Expert LLMs (e.g., Mixtral)
+- Mixture-of-Expert LLMs (e.g., Mixtral, Deepseek-V2 and V3)
 - Embedding Models (e.g. E5-Mistral)
 - Multi-modal LLMs (e.g., LLaVA)
 

docs/source/models/supported_models.md

@@ -137,6 +137,11 @@ See [this page](#generative-models) for more information on how to use generativ
     - :code:`deepseek-ai/DeepSeek-V2`, :code:`deepseek-ai/DeepSeek-V2-Chat` etc.
     -
     - ✅︎
+  * - :code:`DeepseekV3ForCausalLM`
+    - DeepSeek-V3
+    - :code:`deepseek-ai/DeepSeek-V3-Base`, :code:`deepseek-ai/DeepSeek-V3` etc.
+    -
+    - ✅︎
   * - :code:`ExaoneForCausalLM`
     - EXAONE-3
     - :code:`LGAI-EXAONE/EXAONE-3.0-7.8B-Instruct`, etc.
@@ -676,7 +681,7 @@ See [this page](#generative-models) for more information on how to use generativ
     - PaliGemma, PaliGemma 2
     - T + I\ :sup:`E`
     - :code:`google/paligemma-3b-pt-224`, :code:`google/paligemma-3b-mix-224`, :code:`google/paligemma2-3b-ft-docci-448`, etc.
-    - 
+    -
     - ✅︎
     -
   * - :code:`Phi3VForCausalLM`