From 08e6a4e31504ca8cafe9368f3facf18eecd21572 Mon Sep 17 00:00:00 2001
From: iwindy910 <mr.iwindy.zhhh@gmail.com>
Date: Sat, 10 Jan 2026 14:42:16 +0800
Subject: [PATCH] Add files via upload

---
 __init__.py      | 13 ++++++++
 llm_vlm_api.py   | 87 ++++++++++++++++++++++++++++++++++++++++++++++++
 requirements.txt |  3 ++
 3 files changed, 103 insertions(+)
 create mode 100644 __init__.py
 create mode 100644 llm_vlm_api.py
 create mode 100644 requirements.txt

diff --git a/__init__.py b/__init__.py
new file mode 100644
index 0000000..cca6680
--- /dev/null
+++ b/__init__.py
@@ -0,0 +1,13 @@
+from .llm_vlm_api import LLM_VLM_API, LLM_VLM_Prompt
+
+NODE_CLASS_MAPPINGS = {
+    "LLM_VLM_API": LLM_VLM_API,
+    "LLM_VLM_Prompt": LLM_VLM_Prompt
+}
+
+NODE_DISPLAY_NAME_MAPPINGS = {
+    "LLM_VLM_API": "LLM (VLM) API",
+    "LLM_VLM_Prompt": "LLM (VLM) Prompt"
+}
+
+__all__ = ['NODE_CLASS_MAPPINGS', 'NODE_DISPLAY_NAME_MAPPINGS']
\ No newline at end of file
diff --git a/llm_vlm_api.py b/llm_vlm_api.py
new file mode 100644
index 0000000..48d612c
--- /dev/null
+++ b/llm_vlm_api.py
@@ -0,0 +1,87 @@
+import base64
+import numpy as np
+from PIL import Image
+import io as pyio
+from openai import OpenAI
+
+class LLM_VLM_API:
+    @classmethod
+    def INPUT_TYPES(s):
+        return {
+            "required": {
+                "base_url": ("STRING", {"default": ""}),
+                "api_key": ("STRING", {"default": "", "password": True}),
+                "model": ("STRING", {"default": ""}),
+                "temperature": ("FLOAT", {"default": 0.7, "min": 0.0, "max": 2.0, "step": 0.01}),
+            },
+            "optional": {
+                "image": ("IMAGE",),
+                "system prompt": ("STRING", {"forceInput": True}),
+                "text": ("STRING", {"forceInput": True}),
+            }
+        }
+
+    RETURN_TYPES = ("STRING",)
+    RETURN_NAMES = ("response",)
+    FUNCTION = "execute"
+    CATEGORY = "LLM(VLM) API"
+
+    def execute(self, base_url, api_key, model, temperature, **kwargs):
+        system_prompt = kwargs.get("system prompt", "")
+        image = kwargs.get("image", None)
+        text = kwargs.get("text", "")
+
+        if not text and image is None:
+            return ("Error: No text or image provided.",)
+
+        content = []
+        if text:
+            content.append({"type": "text", "text": text})
+
+        if image is not None:
+            img_np = (image[0].cpu().numpy() * 255).astype(np.uint8)
+            pil_img = Image.fromarray(img_np)
+            buf = pyio.BytesIO()
+            pil_img.save(buf, format="PNG")
+            img_b64 = base64.b64encode(buf.getvalue()).decode()
+            content.append({
+                "type": "image_url",
+                "image_url": {"url": f"data:image/png;base64,{img_b64}"}
+            })
+
+        client = OpenAI(api_key=api_key, base_url=base_url)
+
+        try:
+            messages = []
+            if system_prompt:
+                messages.append({"role": "system", "content": system_prompt})
+            
+            messages.append({"role": "user", "content": content})
+
+            resp = client.chat.completions.create(
+                model=model,
+                temperature=temperature,
+                messages=messages
+            )
+            result = resp.choices[0].message.content
+        except Exception as e:
+            result = str(e)
+
+        return (result,)
+
+class LLM_VLM_Prompt:
+    @classmethod
+    def INPUT_TYPES(s):
+        return {
+            "required": {
+                "prompt": ("STRING", {"multiline": True, "default": ""}),
+            }
+        }
+
+    RETURN_TYPES = ("STRING",)
+    RETURN_NAMES = ("text",)
+    FUNCTION = "execute"
+    CATEGORY = "LLM(VLM) API"
+
+    def execute(self, prompt):
+        return (prompt,)
\ No newline at end of file
diff --git a/requirements.txt b/requirements.txt
new file mode 100644
index 0000000..fcd43f4
--- /dev/null
+++ b/requirements.txt
@@ -0,0 +1,3 @@
+openai
+numpy
+Pillow
\ No newline at end of file