Add gr.Workflow app for GLM-5.3 chat via HF Inference Providers

- Bound fn node streams from zai-org/GLM-5.3:zai-org via the HF router
- Visitor OAuth token injected via gr.OAuthToken (billed to the user)
- Pre-wired canvas: Prompt + System Prompt -> GLM-5.3 Chat -> Response

.gitignore

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

app.py

@@ -0,0 +1,43 @@
+from typing import Optional
+
+import gradio as gr
+from huggingface_hub import get_token
+from openai import OpenAI
+
+
+def _client(oauth_token: Optional[gr.OAuthToken]) -> OpenAI:
+    # gr.Workflow injects the visitor's OAuth token into bound functions,
+    # so inference is billed to the user running the workflow. Fall back to
+    # the local `hf auth login` token when there is no OAuth session.
+    token = oauth_token.token if oauth_token else get_token()
+    return OpenAI(
+        base_url="https://router.huggingface.co/v1",
+        api_key=token,
+    )
+
+
+def glm_chat(
+    prompt: str,
+    system_prompt: str = "You are a helpful assistant.",
+    token: Optional[gr.OAuthToken] = None,
+) -> str:
+    """Chat with GLM-5.3 via Hugging Face Inference Providers (streaming)."""
+    stream = _client(token).chat.completions.create(
+        model="zai-org/GLM-5.3:zai-org",
+        messages=[
+            {"role": "system", "content": system_prompt},
+            {"role": "user", "content": prompt},
+        ],
+        stream=True,
+    )
+    return "".join(
+        chunk.choices[0].delta.content or ""
+        for chunk in stream
+        if chunk.choices and chunk.choices[0].delta
+    )
+
+
+gr.Workflow(
+    bind={"GLM-5.3 Chat": glm_chat},
+    graph="workflow.json",
+).launch()

requirements.txt

@@ -0,0 +1,3 @@
+gradio>=6.26.0
+openai
+huggingface_hub

workflow.json

@@ -1 +1 @@
-{"schema_version":"2","name":"My Workflow","runtime":{"default":"client"},"references":[],"operators":[],"subjects":[],"edges":[],"view":{"default":"canvas"}}
+{"schema_version":"2","name":"GLM-5.3 Chat Workflow","references":[{"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},{"id":"ref_system","label":"System 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_glm","label":"GLM-5.3 Chat","role":"operator","kind":"fn","fn":"GLM-5.3 Chat","inputs":[{"id":"prompt","label":"Prompt","type":"text","required":true},{"id":"system_prompt","label":"System Prompt","type":"text"}],"outputs":[{"id":"out_0","label":"Response","type":"text","output_index":0}],"data":{},"width":200}],"subjects":[{"id":"sub_response","label":"Response","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_prompt","from_port_id":"out","to_node_id":"op_glm","to_port_id":"prompt","type":"text"},{"id":"e2","from_node_id":"ref_system","from_port_id":"out","to_node_id":"op_glm","to_port_id":"system_prompt","type":"text"},{"id":"e3","from_node_id":"op_glm","from_port_id":"out_0","to_node_id":"sub_response","to_port_id":"in","type":"text"}]}