Fix prompt expansion to handle multiple variations and enforce length limit c0b9c7a Zoro-chi commited on Apr 17
Fix prompt expansion with better text processing and robust fallbacks cf7ae54 Zoro-chi commited on Apr 17
Limit prompt expansion to maximum 40 words with strict focus on original prompt 448eec3 Zoro-chi commited on Apr 17
Completely revise GGUF loading to match direct ctransformers example 4547a45 Zoro-chi commited on Apr 17
Fix GGUF model loading using correct configuration parameters format b367fc7 Zoro-chi commited on Apr 17
Implement structured creative prompt instructions for FLAN-T5-base model dac04d5 Zoro-chi commited on Apr 17
Switch to google/flan-t5-base and improve prompt expansion for better quality b26638e Zoro-chi commited on Apr 17
Enhance prompt expansion for FLAN-T5 model with specialized seq2seq handling 590186e Zoro-chi commited on Apr 17
Add support for google/flan-t5-small with proper sequence-to-sequence model handling 34d21a4 Zoro-chi commited on Apr 17
Fix chat template error with DistilGPT2 by adding support for non-chat models b83f7c0 Zoro-chi commited on Apr 17
Switch to TinyLlama GGUF model for much faster inference in Hugging Face Spaces 37ed2a0 Zoro-chi commited on Apr 17
Fix LLM usage in Hugging Face Spaces with smaller model and import handling f1cc7f4 Zoro-chi commited on Apr 17
Fix pipeline initialization error and improve LLM handling for Hugging Face Spaces a018b91 Zoro-chi commited on Apr 17
Fix Hugging Face Spaces issues: LLM URL error and app ID connection problems e366aec Zoro-chi commited on Apr 17
