From 1731f41a03395e49cd8eedbe3ccb59a796fdf856 Mon Sep 17 00:00:00 2001 From: Mikyo King Date: Mon, 8 Jan 2024 14:38:34 -0700 Subject: [PATCH] docs: move spec to openinference --- .gitignore | 1 + README.md | 6 +++++- trace/README.md | 4 ++++ 3 files changed, 10 insertions(+), 1 deletion(-) create mode 100644 .gitignore diff --git a/.gitignore b/.gitignore new file mode 100644 index 0000000..c2658d7 --- /dev/null +++ b/.gitignore @@ -0,0 +1 @@ +node_modules/ diff --git a/README.md b/README.md index 0f00ccf..871e6d9 100644 --- a/README.md +++ b/README.md @@ -1,9 +1,13 @@ # OpenInference +> [!IMPORTANT] +> OpenInference Tracing is now being actively worked on via: [OpenInference](https://github.com/Arize-ai/openinference) +> This repository is maintained for historical purposes only. + This is a a working draft of the specification for OpenInference - a specification that encompass two data models: - [Inferences](./inference/README.md) -- [Traces](./trace/README.md) +- [Traces](https://github.com/Arize-ai/openinference/tree/main/spec) Inferences are designed to capture inference logs from a variety of model types and use-cases where as traces are designed to capture the execution of an application that results in invocations of an LLM. diff --git a/trace/README.md b/trace/README.md index 39c90d6..3582827 100644 --- a/trace/README.md +++ b/trace/README.md @@ -1,5 +1,9 @@ # OpenInference Tracing +> [!IMPORTANT] +> OpenInference Tracing is now being actively worked on via: [OpenInference](https://github.com/Arize-ai/openinference) +> This repository is maintained for historical purposes only. + The OpenInference Tracing specification is edited in markdown files found in the [spec directory](./spec/README.md). OpenInference Tracing is a specification for capturing and storing LLM application executions. It's designed to provide insight into the invocation of LLMs and the surrounding application context such as retrieval from vector stores and the usage of external tools such as search engines or APIs. The specification is transport and file-format agnostic, and is intended to be used in conjunction with other specifications such as JSON, ProtoBuf, and DataFrames.