v0.1.3
Pre-release
Pre-release
Full Changelog: https://github.com/RobinQu/instinct.cpp/commits/v0.1.3
Sorry for the delay. Hacking with new agent executor costs me more time than expected. But it's fun and enlightening. As a result, work on rerank models will delay to next release.
- Features
- New
LLMCompilerAgentExecutor
as default executor inmini-assistant
. Please see this notebook for evaluation results. POST /v1/threads/runs
endpoint, which is missed in last release, is added.xn::steps::branch
andBranchStepFunction
for binary branch composing.
- New
- Improvements
- Migrate
instinct-lab
to standalone repo. Benchmarks and experimental code will be moved there. - Many fixes for typos in
mini-assistant
API endpoints. - Add more trace logs in debug mode.
BaseChatModel
andBaseLLM
supports runtime configuration withconfigure(const ModelOverrides&)
method.HttpLibServer
defaults to listen0.0.0.0
.
- Migrate
SHA256 digest for assests:
e0ebb75722e6eedba6c60b01a1a1938e15bd1ea043949c0e8cc97ee60fedc099 mini-assistant_linux-amd64.tar.gz
114a04e95dee7b51559fe78003c78dfd996ee2d5e035e5881f6fa16f9a6acf25 mini-assistant_macos-arm64.tar.gz
9db7fb3513b0546ce2cfaea8871a9df7437d6cbf8e8ac57edf25b4283b656cca doc-agent_linux-amd64.tar.gz
90385ddb9f3e646eda3b05c0e9b46af4cbb5b49b598bb391dff322d6172097d1 doc-agent_macos-arm64.tar.gz