From bec4ede3f67cf263547eebeeb53fa4af443d0a1c Mon Sep 17 00:00:00 2001 From: Harjot Gill Date: Sun, 19 Jan 2025 18:47:38 -0800 Subject: [PATCH] self-hosted: support o1 model (#203) --- docs/self-hosted/azure-devops.md | 12 ++++-------- docs/self-hosted/bitbucket.md | 16 ++++++---------- docs/self-hosted/github.md | 10 +++------- docs/self-hosted/gitlab.md | 12 ++++-------- 4 files changed, 17 insertions(+), 33 deletions(-) diff --git a/docs/self-hosted/azure-devops.md b/docs/self-hosted/azure-devops.md index 980a24f..0468100 100644 --- a/docs/self-hosted/azure-devops.md +++ b/docs/self-hosted/azure-devops.md @@ -11,7 +11,7 @@ The self-hosted option is only available for CodeRabbit Enterprise customers wit ::: -## Create a Azure DevOps User +## Create an Azure DevOps User - **Username**: Set the username to "CodeRabbit" for easier identification (optional). - **Profile Image**: Use the CodeRabbitAI logo for the user image (optional). @@ -61,14 +61,10 @@ LLM_PROVIDER=azure-openai LLM_TIMEOUT=360000 AZURE_OPENAI_ENDPOINT= AZURE_OPENAI_API_KEY= -## it is recommended to use gpt-4o-mini, o1-mini, and o1-preview deployments +## it is recommended to use gpt-4o-mini, o1-mini, and o1-preview deployments. The production release of o1 model is inferior to the preview release as of now. Also, please make sure that the deployment name of o1-preview mentions "o1-preview" in it. AZURE_GPT4OMINI_DEPLOYMENT_NAME= AZURE_O1MINI_DEPLOYMENT_NAME=[] -AZURE_O1_DEPLOYMENT_NAME=[] -## gpt-4o is optional -AZURE_GPT4O_DEPLOYMENT_NAME= -## gpt-4-turbo is optional: it’s expensive but provides better reviews than gpt-4o -AZURE_GPT4TURBO_DEPLOYMENT_NAME=[] +AZURE_O1_DEPLOYMENT_NAME=[] # if using AWS Bedrock AWS_ACCESS_KEY_ID= @@ -115,7 +111,7 @@ docker pull /coderabbit-agent:latest ### Verify the image is up -You can query `/health` endpoint to verify that the coderabbit-agent service is up and running. +You can query `/health` endpoint to verify that the `coderabbit-agent` service is up and running. ```bash curl 127.0.0.1:8080/health diff --git a/docs/self-hosted/bitbucket.md b/docs/self-hosted/bitbucket.md index bf5f45f..a2879c6 100644 --- a/docs/self-hosted/bitbucket.md +++ b/docs/self-hosted/bitbucket.md @@ -1,5 +1,5 @@ --- -title: Use Self-Hosted CodeRabbit With Bitbucket Datacenter +title: Use Self-Hosted CodeRabbit With Bitbucket Datacenter sidebar_label: Bitbucket Datacenter description: Instructions to self-host CodeRabbit and integrate it with Bitbucket Datacenter. sidebar_position: 4 @@ -36,9 +36,9 @@ Generate a personal access token for the CodeRabbit user to be added in the `.en - "Comment Added" 3. **Add Webhook URL**: Enter the URL pointing to the CodeRabbit service, followed by `/bitbucket_server_webhooks` (e.g., `http://127.0.0.1:8080/bitbucket_server_webhooks`). -## Prepare an `.env` file +## Prepare a `.env` file -Create an `.env` file with the following content: +Create a `.env` file with the following content: ```bash # if using OpenAI @@ -54,14 +54,10 @@ LLM_PROVIDER=azure-openai LLM_TIMEOUT=360000 AZURE_OPENAI_ENDPOINT= AZURE_OPENAI_API_KEY= -## it is recommended to use gpt-4o-mini, o1-mini, and o1-preview deployments +## it is recommended to use gpt-4o-mini, o1-mini, and o1-preview deployments. The production release of o1 model is inferior to the preview release as of now. Also, please make sure that the deployment name of o1-preview mentions "o1-preview" in it. AZURE_GPT4OMINI_DEPLOYMENT_NAME= AZURE_O1MINI_DEPLOYMENT_NAME=[] -AZURE_O1_DEPLOYMENT_NAME=[] -## gpt-4o is optional -AZURE_GPT4O_DEPLOYMENT_NAME= -## gpt-4-turbo is optional: it’s expensive but provides better reviews than gpt-4o -AZURE_GPT4TURBO_DEPLOYMENT_NAME=[] +AZURE_O1_DEPLOYMENT_NAME=[] # if using AWS Bedrock AWS_ACCESS_KEY_ID= @@ -108,7 +104,7 @@ docker pull us-docker.pkg.dev/coderabbitprod/self-hosted/coderabbit-agent:latest ### Verify the image is up -You can query `/health` endpoint to verify that the coderabbit-agent service is up and running. +You can query `/health` endpoint to verify that the `coderabbit-agent` service is up and running. ```bash curl 127.0.0.1:8080/health diff --git a/docs/self-hosted/github.md b/docs/self-hosted/github.md index 0c3937c..7bb7ed5 100644 --- a/docs/self-hosted/github.md +++ b/docs/self-hosted/github.md @@ -65,14 +65,10 @@ LLM_PROVIDER=azure-openai LLM_TIMEOUT=360000 AZURE_OPENAI_ENDPOINT= AZURE_OPENAI_API_KEY= -## it is recommended to use gpt-4o-mini, o1-mini, and o1-preview deployments +## it is recommended to use gpt-4o-mini, o1-mini, and o1-preview deployments. The production release of o1 model is inferior to the preview release as of now. Also, please make sure that the deployment name of o1-preview mentions "o1-preview" in it. AZURE_GPT4OMINI_DEPLOYMENT_NAME= AZURE_O1MINI_DEPLOYMENT_NAME=[] -AZURE_O1_DEPLOYMENT_NAME=[] -## gpt-4o is optional -AZURE_GPT4O_DEPLOYMENT_NAME= -## gpt-4-turbo is optional: it’s expensive but provides better reviews than gpt-4o -AZURE_GPT4TURBO_DEPLOYMENT_NAME=[] +AZURE_O1_DEPLOYMENT_NAME=[] # if using AWS Bedrock AWS_ACCESS_KEY_ID= @@ -145,7 +141,7 @@ docker run --env-file .env --publish 127.0.0.1:8080:8080 /coder ### Verify the image is up -You can query `/health` endpoint to verify that the coderabbit-agent service is up and running. +You can query `/health` endpoint to verify that the `coderabbit-agent` service is up and running. ```bash curl 127.0.0.1:8080/health diff --git a/docs/self-hosted/gitlab.md b/docs/self-hosted/gitlab.md index 55fefae..ae402eb 100644 --- a/docs/self-hosted/gitlab.md +++ b/docs/self-hosted/gitlab.md @@ -42,9 +42,9 @@ Consult official CodeRabbitAI documentation for a detailed [guide](https://docs. - Issues events - Merge request events -## Prepare an `.env` file +## Prepare a `.env` file -Create an `.env` file with the following content: +Create a `.env` file with the following content: ```bash # if using OpenAI @@ -60,14 +60,10 @@ LLM_PROVIDER=azure-openai LLM_TIMEOUT=360000 AZURE_OPENAI_ENDPOINT= AZURE_OPENAI_API_KEY= -## it is recommended to use gpt-4o-mini, o1-mini, and o1-preview deployments +## it is recommended to use gpt-4o-mini, o1-mini, and o1-preview deployments. The production release of o1 model is inferior to the preview release as of now. Also, please make sure that the deployment name of o1-preview mentions "o1-preview" in it. AZURE_GPT4OMINI_DEPLOYMENT_NAME= AZURE_O1MINI_DEPLOYMENT_NAME=[] -AZURE_O1_DEPLOYMENT_NAME=[] -## gpt-4o is optional -AZURE_GPT4O_DEPLOYMENT_NAME= -## gpt-4-turbo is optional: it’s expensive but provides better reviews than gpt-4o -AZURE_GPT4TURBO_DEPLOYMENT_NAME=[] +AZURE_O1_DEPLOYMENT_NAME=[] # if using AWS Bedrock AWS_ACCESS_KEY_ID=