Actual text streaming (#4)

- Actual text streaming (77430551fc21bcd4ec677ed81c2a8a8564b42612)


Co-authored-by: Charles Bensimon <[email protected]>

app.py

@@ -76,8 +76,8 @@ def model_inference(
 
     thread = Thread(target=model.generate, kwargs=generation_args)
     thread.start()
-    thread.join()
-    
+
+    yield "..."
     buffer = ""