add retry logic for nvidia api errors

pass model explicitly to all pipeline stages

add llama

fix gemini model

force rebuild

fix gemini model name and remove models/ prefix

fix gemini model prefix issue

use gemini-1.5-flash-latest

use gemini-2.0-flash-lite

Merge branch 'main' of https://huggingface.co/spaces/srishtichugh/blog2code-api

initial commit

initial commit