Rename vLLM app to vllm-inference and use dynamic URL resolution

Rewrite transcribe_generator.py to match Cohere quickstart logic

Fix Cohere transcription: correct model class, load_audio, device_map, inputs pattern

to make sure of image generation observability : add an explicit flush right before the script ends (#3)

Update backend tracing of crewai with litellm (#2)

pass uuid for session tracking

update tracking with sessions

final update to match frontent

nice look update

Update crew2.py

Update crew2.py

Update app.py

Update requirements.txt

Update requirements.txt

Update requirements.txt

Update requirements.txt

Update requirements.txt

Update requirements.txt

Update requirements.txt

Update requirements.txt

Update requirements.txt

Update requirements.txt

Update requirements.txt

Update requirements.txt

Update app.py

Update requirements.txt

Update requirements.txt

Create crew2.py

Create requirements.txt

Create app.py

initial commit