LlamaEdge-RAG 0.3.2

Major changes:

• Adapt to llama-core v0.7.1
• Improve the log messages

LlamaEdge-RAG 0.3.1

Major changes:

• Support new models:
  • second-state/gemma-1.1-7b-it-GGUF
  • second-state/gemma-1.1-2b-it-GGUF
  • second-state/CodeGemma-7b-it-GGUF
  • second-state/Octopus-v2-GGUF

LlamaEdge-RAG 0.3.0

Major changes:

• (BREAKING) Update CLI options

  • Keep RAG related CLI options
  • Replace --system-prompt with --rag-prompt option
  • Add --chunk-capacity option

• Adapt to llama-core v0.8.0

LlamaEdge-RAG 0.2.0

Major changes:

• Rename /v1/rag/embeddings to /v1/create/rag
• Remove /v1/rag/query
• Update README

LlamaEdge-RAG 0.1.0

Major features:

• /v1/rag/embeddings endpoint as an all-in-one way to convert a *.txt or*.md document to embeddings.
• /v1/rag/query endpoint as an equivalent of /v1/chat/completions for chatting with RAG models.
• rag-api-server.wasm for HTTP scheme only.
• rag-api-server-full.wasm for both HTTP and HTTPS schemes.