DeepSeek-V4.1-Flash image description workflow (gradio 6.26.0)

.gitignore

@@ -0,0 +1,2 @@
+.venv/
+__pycache__/

README.md

@@ -4,7 +4,7 @@ emoji: ⚡
 colorFrom: indigo
 colorTo: indigo
 sdk: gradio
-sdk_version: 6.25.0
+sdk_version: 6.26.0
 app_file: run.py
 pinned: false
 hf_oauth: true

requirements.txt

@@ -0,0 +1 @@
+openai

run.py

@@ -1,6 +1,36 @@
+import os
 import gradio as gr
+from openai import OpenAI
 
-demo = gr.Workflow()
+client = OpenAI(
+    base_url="https://router.huggingface.co/v1",
+    api_key=os.environ["HF_TOKEN"],
+)
 
-if __name__ == "__main__":
-    demo.launch()
+
+def describe_image(image_url: str, prompt: str) -> str:
+    """Stream a one-sentence description of an image from DeepSeek-V4.1-Flash."""
+    stream = client.chat.completions.create(
+        model="deepseek-ai/DeepSeek-V4.1-Flash:novita",
+        messages=[
+            {
+                "role": "user",
+                "content": [
+                    {"type": "text", "text": prompt},
+                    {"type": "image_url", "image_url": {"url": image_url}},
+                ],
+            }
+        ],
+        stream=True,
+    )
+    output = ""
+    for chunk in stream:
+        if chunk.choices and chunk.choices[0].delta.content:
+            output += chunk.choices[0].delta.content
+    return output
+
+
+gr.Workflow(
+    bind={"Describe Image": describe_image},
+    graph="workflow.json",
+).launch()

workflow.json

@@ -1 +1 @@
-{"schema_version":"2","name":"My Workflow","runtime":{"default":"client"},"references":[],"operators":[],"subjects":[],"edges":[],"view":{"default":"canvas"}}
+{"schema_version":"2","name":"DeepSeek Image Describer","references":[{"id":"ref_image_url","label":"Image URL","role":"reference","asset_type":"text","inputs":[{"id":"in","label":"Text","type":"text"}],"outputs":[{"id":"out","label":"Text","type":"text"}],"data":{"out":""},"width":200},{"id":"ref_prompt","label":"Prompt","role":"reference","asset_type":"text","inputs":[{"id":"in","label":"Text","type":"text"}],"outputs":[{"id":"out","label":"Text","type":"text"}],"data":{"out":""},"width":200}],"operators":[{"id":"op_describe","label":"Describe Image","role":"operator","kind":"fn","fn":"Describe Image","inputs":[{"id":"in_image_url","label":"image_url","type":"text","required":true},{"id":"in_prompt","label":"prompt","type":"text","required":true}],"outputs":[{"id":"out_0","label":"output","type":"text","output_index":0}],"data":{},"width":200}],"subjects":[{"id":"sub_description","label":"Description","role":"subject","asset_type":"text","inputs":[{"id":"in","label":"Text","type":"text"}],"outputs":[{"id":"out","label":"Text","type":"text"}],"data":{"in":""},"width":200}],"edges":[{"id":"e1","from_node_id":"ref_image_url","from_port_id":"out","to_node_id":"op_describe","to_port_id":"in_image_url","type":"text"},{"id":"e2","from_node_id":"ref_prompt","from_port_id":"out","to_node_id":"op_describe","to_port_id":"in_prompt","type":"text"},{"id":"e3","from_node_id":"op_describe","from_port_id":"out_0","to_node_id":"sub_description","to_port_id":"in","type":"text"}]}