- Fixes a bug where errors weren't being propagated through the stack correctly (manifested in #4).
LlamaRunner.RunState.failed()
is now created with an optionalError
— this should be non-nil but because of the bridging with Objective-C/C++ there may be cases where this is not the case.
llama.swift → CameLLM:
- This is a maintenance update for llama.swift; all future feature development is happening at 🐪 CameLLM