Fix DynamicCache error: disable use_cache to avoid 'seen_tokens' attribute error 1dac243 BelikanM commited on 22 days ago
Enhance chat: RAG integration + fluent token streaming + .dat data context 736128c BelikanM commited on 22 days ago
Fix BitsAndBytes unavailable on HF Spaces: adaptive quantization with CPU optimizations da9f45b BelikanM commited on 22 days ago
Fix Phi-3 performance: attn_implementation=eager, optimized generation params, single thread CPU e196a12 BelikanM commited on 22 days ago
Optimize Phi-3 quantifier 4-bit + CLIP streaming tokens for ultra-fast generation b73de9c BelikanM commited on 22 days ago