vamseeachanta Streaming Responses
Streaming Responses
For better UX with long answers:
def query_streaming(self, question, top_k=5):
"""Stream RAG response for real-time display."""
context = self.get_context(question, top_k)
prompt = self.build_prompt(context, question)
# Anthropic streaming
with anthropic.Anthropic().messages.stream(
model="Codex-sonnet-4-6",
max_tokens=1024,
messages=[{"role": "user", "content": prompt}]
) as stream:
for text in stream.text_stream:
yield text