fix: Update default chat model to grok-4-1-fast-reasoning and enhance logging for LLM responses

This commit is contained in:
bsiggel
2026-03-19 09:50:31 +00:00
parent 7fffdb2660
commit 2ac83df1e0
2 changed files with 15 additions and 3 deletions

View File

@@ -317,6 +317,18 @@ async def invoke_and_format_response(
"total_tokens": getattr(u, 'input_tokens', 0) + getattr(u, 'output_tokens', 0)
}
# Log complete LLM response
ctx.logger.info("=" * 80)
ctx.logger.info("📤 LLM RESPONSE")
ctx.logger.info("-" * 80)
ctx.logger.info(f"Model: {model_name}")
ctx.logger.info(f"Completion ID: {completion_id}")
ctx.logger.info(f"Usage: {usage['prompt_tokens']} prompt + {usage['completion_tokens']} completion = {usage['total_tokens']} total tokens")
ctx.logger.info("-" * 80)
ctx.logger.info("Content:")
ctx.logger.info(content)
ctx.logger.info("=" * 80)
# Format OpenAI-compatible response
response_body = {
'id': completion_id,