Releases: LlamaEdge/rag-api-server
Releases · LlamaEdge/rag-api-server
LlamaEdge-RAG 0.13.15
Major changes:
- Upgrade
llama-core
dep to0.30.0
- Upgrade
chat-prompts
dep to0.26.1
LlamaEdge-RAG 0.13.14
Major changes:
- Support Llama-4 (text-only)
- Verified on
ggml plugin b5074
LlamaEdge-RAG 0.13.13
Major changes:
- Support Tessa-T1 react reasoning models
LlamaEdge-RAG 0.13.12
Major changes:
- Support LGAI-EXAONE/EXAONE-3.5-Instruct models
LlamaEdge-RAG 0.13.11
Major changes:
-
Support LGAI-EXAONE/EXAONE-Deep models
-
Verified on
ggml plugin b4917
LlamaEdge-RAG 0.13.10
Major changes:
- Automatically switch
rag_policy
fromsystem-message
tolast-user-message
when the model does not support system prompt.
LlamaEdge-RAG 0.13.9
Major changes:
- Upgrade
llama-core
dep to0.28.2
- Verified on
ggml plugin b4875
LlamaEdge-RAG 0.13.8
Major changes:
- Upgrade
llama-core
dep to0.28.0
- Upgrade
chat-prompts
dep to0.21.2
- Upgrade
endpoints
dep to0.25.1
LlamaEdge-RAG 0.13.7
Major change:
- Bug fix
LlamaEdge-RAG 0.13.6
Major changes:
- (NEW) Add
--include-usage
CLI option. Note that theinclude_usage
param in requests has higher priority than the--include-usage
CLI option.