Cypheros-de/Delphi11LlamaCppBindings
Delphi 11+ bindings for llama.cpp (b9050+) with full GPU acceleration. Updated from the original Embarcadero fork: new memory API, vocab object, backend loader, CUDA 13, RTX 30xx/40xx/50xx support, Flash Attention, quantized KV cache, Jinja2 chat templates, and multimodal vision inference via the mtmd API.
GitHub repository with 5 stars and 1 forks.
Language: Pascal