Update app.py
Browse files
app.py
CHANGED
|
@@ -4,12 +4,10 @@ from huggingface_hub import InferenceClient
|
|
| 4 |
import requests
|
| 5 |
|
| 6 |
url = "https://huggingface.co/models-json?num_parameters=min%3A128B&inference_provider=cerebras%2Cnebius%2Cfireworks-ai%2Ctogether%2Csambanova%2Cnovita%2Cgroq%2Cnscale%2Chyperbolic%2Cfeatherless-ai%2Cfal-ai%2Ccohere%2Creplicate%2Chf-inference&sort=trending&withCount=true"
|
| 7 |
-
|
| 8 |
res = requests.get(url).json()
|
| 9 |
model_ids = [m['id'] for m in res.get('models', [])]
|
| 10 |
|
| 11 |
client = InferenceClient(provider="novita", api_key=os.environ["HF_TOKEN"])
|
| 12 |
-
#models = ["deepseek-ai/DeepSeek-V3.1","zai-org/GLM-4.5","Qwen/Qwen3-Coder-480B-A35B-Instruct", "Qwen/Qwen3-Coder-480B-A35B-Instruct-FP8","deepseek-ai/DeepSeek-R1","deepseek-ai/DeepSeek-R1-0528","Qwen/Qwen3-Next-80B-A3B-Thinking"]
|
| 13 |
|
| 14 |
def respond(message, history, model):
|
| 15 |
history.append((message, ""))
|
|
@@ -20,8 +18,9 @@ def respond(message, history, model):
|
|
| 20 |
messages=[{"role": "user", "content": message}],
|
| 21 |
stream=True
|
| 22 |
):
|
| 23 |
-
|
| 24 |
-
|
|
|
|
| 25 |
history[-1] = (message, full_reply)
|
| 26 |
yield "", history
|
| 27 |
|
|
@@ -29,7 +28,7 @@ with gr.Blocks() as demo:
|
|
| 29 |
gr.Markdown("## AI")
|
| 30 |
chatbot = gr.Chatbot(height=400)
|
| 31 |
msg = gr.Textbox(label="Ask me smth")
|
| 32 |
-
model_dd = gr.Dropdown(model_ids, label="Model", value=
|
| 33 |
with gr.Row():
|
| 34 |
submit_btn = gr.Button("Send")
|
| 35 |
clear = gr.Button("Clear")
|
|
|
|
| 4 |
import requests
|
| 5 |
|
| 6 |
url = "https://huggingface.co/models-json?num_parameters=min%3A128B&inference_provider=cerebras%2Cnebius%2Cfireworks-ai%2Ctogether%2Csambanova%2Cnovita%2Cgroq%2Cnscale%2Chyperbolic%2Cfeatherless-ai%2Cfal-ai%2Ccohere%2Creplicate%2Chf-inference&sort=trending&withCount=true"
|
|
|
|
| 7 |
res = requests.get(url).json()
|
| 8 |
model_ids = [m['id'] for m in res.get('models', [])]
|
| 9 |
|
| 10 |
client = InferenceClient(provider="novita", api_key=os.environ["HF_TOKEN"])
|
|
|
|
| 11 |
|
| 12 |
def respond(message, history, model):
|
| 13 |
history.append((message, ""))
|
|
|
|
| 18 |
messages=[{"role": "user", "content": message}],
|
| 19 |
stream=True
|
| 20 |
):
|
| 21 |
+
content = chunk.choices[0].delta.content
|
| 22 |
+
if content:
|
| 23 |
+
full_reply += content
|
| 24 |
history[-1] = (message, full_reply)
|
| 25 |
yield "", history
|
| 26 |
|
|
|
|
| 28 |
gr.Markdown("## AI")
|
| 29 |
chatbot = gr.Chatbot(height=400)
|
| 30 |
msg = gr.Textbox(label="Ask me smth")
|
| 31 |
+
model_dd = gr.Dropdown(model_ids, label="Model", value=model_ids[0] if model_ids else "")
|
| 32 |
with gr.Row():
|
| 33 |
submit_btn = gr.Button("Send")
|
| 34 |
clear = gr.Button("Clear")
|