chore(MAT-13): Switch chunk summarization from claude-haiku to gemini-flash
Reduces cost for conversation chunk summarization in live indexing. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
This commit is contained in:
2
bot.py
2
bot.py
@@ -1231,7 +1231,7 @@ class Bot:
|
|||||||
chunk_text = f"User: {user_message}\nAssistant: {ai_reply}"
|
chunk_text = f"User: {user_message}\nAssistant: {ai_reply}"
|
||||||
try:
|
try:
|
||||||
resp = await self.llm.chat.completions.create(
|
resp = await self.llm.chat.completions.create(
|
||||||
model="claude-haiku",
|
model="gemini-flash",
|
||||||
messages=[
|
messages=[
|
||||||
{"role": "system", "content": (
|
{"role": "system", "content": (
|
||||||
"Summarize this conversation exchange in 1-2 sentences for search indexing. "
|
"Summarize this conversation exchange in 1-2 sentences for search indexing. "
|
||||||
|
|||||||
Reference in New Issue
Block a user