Switch to Groq API with llama-3.1-8b-instant

Use Together AI provider with Llama-3.2-3B-Instruct

Back to InferenceClient with Mixtral-8x7B

Fix syntax error in app.py

Switch to direct HF API calls with zephyr-7b-beta

Use hf-inference provider with Qwen2.5-72B-Instruct

Switch to Meta-Llama-3.1-8B-Instruct for chat completion

Fix: use chat_completion instead of text_generation

Switch to HF Inference API for better recipe generation

Improve recipe output length and quality

Fix: load flan-t5 directly instead of pipeline

Switch to flan-t5-base for fast CPU inference

Deploy recipe generator to HF Spaces

Initial Streamlit recipe app