docs: fix deployment commands — vLLM requires --language-model-only (text-only weights); transformers needs accelerate and drop --continuous-batching; SGLang note

README.md

@@ -207,6 +207,8 @@ uv pip install sglang[all]
 
 The following will create API endpoints at `http://localhost:8000/v1`:
 
+> **Note:** This open-weight release ships only the language-model weights (no vision tower). If your SGLang version attempts to build the multimodal/vision components at load time, startup may fail on missing vision weights; in that case, run with the version's text/language-model-only option (see `python -m sglang.launch_server --help`).
+
 - Standard Version: The following command can be used to create an API endpoint with maximum context length 262,144 tokens using tensor parallel on 8 GPUs.
    ```shell
    python -m sglang.launch_server \
@@ -239,13 +241,16 @@ uv pip install vllm --torch-backend=auto
 
 The following will create API endpoints at `http://localhost:8000/v1`:
 
+> **Note:** This open-weight release ships only the language-model weights, so the `--language-model-only` flag is **required**. It tells vLLM to skip the vision encoder and multimodal profiling; without it, vLLM attempts to initialize vision-tower weights that are not present in the checkpoint and startup fails.
+
 - Standard Version: The following command can be used to create an API endpoint with maximum context length 262,144 tokens using tensor parallel on 8 GPUs.
    ```shell
    vllm serve Kwaipilot/KAT-Coder-V2.5-Dev \
        --port 8000 \
        --tensor-parallel-size 8 \
        --max-model-len 262144 \
-       --reasoning-parser qwen3
+       --reasoning-parser qwen3 \
+       --language-model-only
    ```
 - Tool Call: To support tool use, you can use the following command.
    ```shell
@@ -255,15 +260,7 @@ The following will create API endpoints at `http://localhost:8000/v1`:
        --max-model-len 262144 \
        --reasoning-parser qwen3 \
        --enable-auto-tool-choice \
-       --tool-call-parser qwen3_coder
-   ```
-- Text-Only: The following command skips the vision encoder and multimodal profiling to free up memory for additional KV cache:
-   ```shell
-   vllm serve Kwaipilot/KAT-Coder-V2.5-Dev \
-       --port 8000 \
-       --tensor-parallel-size 8 \
-       --max-model-len 262144 \
-       --reasoning-parser qwen3 \
+       --tool-call-parser qwen3_coder \
        --language-model-only
    ```
 
@@ -273,16 +270,16 @@ KTransformers is a flexible framework for experiencing cutting-edge LLM inferenc
 
 #### Hugging Face Transformers
 
-Hugging Face Transformers contains a lightweight server which can be used for quick testing and moderate load deployment. The latest transformers is required for KAT-Coder-V2.5-Dev:
+Hugging Face Transformers contains a lightweight server which can be used for quick testing and moderate load deployment. The latest transformers is required for KAT-Coder-V2.5-Dev. Installing `accelerate` is also required for multi-GPU (sharded) loading:
 
 ```shell
-pip install "transformers[serving]"
+pip install "transformers[serving]" accelerate
 ```
 
 Then, run transformers serve to launch a server with API endpoints at `http://localhost:8000/v1`; it will place the model on accelerators if available:
 
 ```shell
-transformers serve Kwaipilot/KAT-Coder-V2.5-Dev --port 8000 --continuous-batching
+transformers serve Kwaipilot/KAT-Coder-V2.5-Dev --port 8000
 ```
 
 ## Using KAT-Coder-V2.5-Dev via the Chat Completions API