Skip to content

Commit bec4ede

Browse files
harjotgillharjotgill
authored
self-hosted: support o1 model (#203)
1 parent 4f3783c commit bec4ede

File tree

4 files changed

+17
-33
lines changed

4 files changed

+17
-33
lines changed

Diff for: ‎docs/self-hosted/azure-devops.md

+4-8
Original file line numberDiff line numberDiff line change
@@ -11,7 +11,7 @@ The self-hosted option is only available for CodeRabbit Enterprise customers wit
1111

1212
:::
1313

14-
## Create a Azure DevOps User
14+
## Create an Azure DevOps User
1515

1616
- **Username**: Set the username to "CodeRabbit" for easier identification (optional).
1717
- **Profile Image**: Use the CodeRabbitAI logo for the user image (optional).
@@ -61,14 +61,10 @@ LLM_PROVIDER=azure-openai
6161
LLM_TIMEOUT=360000
6262
AZURE_OPENAI_ENDPOINT=<azure-openai-endpoint>
6363
AZURE_OPENAI_API_KEY=<key>
64-
## it is recommended to use gpt-4o-mini, o1-mini, and o1-preview deployments
64+
## 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.
6565
AZURE_GPT4OMINI_DEPLOYMENT_NAME=<gpt-4o-mini-deployment-name>
6666
AZURE_O1MINI_DEPLOYMENT_NAME=[<o1-mini-deployment-name>]
67-
AZURE_O1_DEPLOYMENT_NAME=[<o1-preview-deployment-name>]
68-
## gpt-4o is optional
69-
AZURE_GPT4O_DEPLOYMENT_NAME=<gpt-4o-deployment-name, modelVersion: 2024-08-06>
70-
## gpt-4-turbo is optional: it’s expensive but provides better reviews than gpt-4o
71-
AZURE_GPT4TURBO_DEPLOYMENT_NAME=[<gpt-4-turbo-deployment-name, modelVersion: turbo-2024-04-09>]
67+
AZURE_O1_DEPLOYMENT_NAME=[<o1-deployment-name>]
7268

7369
# if using AWS Bedrock
7470
AWS_ACCESS_KEY_ID=<aws-access-key>
@@ -115,7 +111,7 @@ docker pull <docker-registry>/coderabbit-agent:latest
115111

116112
### Verify the image is up
117113

118-
You can query `/health` endpoint to verify that the coderabbit-agent service is up and running.
114+
You can query `/health` endpoint to verify that the `coderabbit-agent` service is up and running.
119115

120116
```bash
121117
curl 127.0.0.1:8080/health

Diff for: ‎docs/self-hosted/bitbucket.md

+6-10
Original file line numberDiff line numberDiff line change
@@ -1,5 +1,5 @@
11
---
2-
title: Use Self-Hosted CodeRabbit With Bitbucket Datacenter
2+
title: Use Self-Hosted CodeRabbit With Bitbucket Datacenter
33
sidebar_label: Bitbucket Datacenter
44
description: Instructions to self-host CodeRabbit and integrate it with Bitbucket Datacenter.
55
sidebar_position: 4
@@ -36,9 +36,9 @@ Generate a personal access token for the CodeRabbit user to be added in the `.en
3636
- "Comment Added"
3737
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`).
3838

39-
## Prepare an `.env` file
39+
## Prepare a `.env` file
4040

41-
Create an `.env` file with the following content:
41+
Create a `.env` file with the following content:
4242

4343
```bash
4444
# if using OpenAI
@@ -54,14 +54,10 @@ LLM_PROVIDER=azure-openai
5454
LLM_TIMEOUT=360000
5555
AZURE_OPENAI_ENDPOINT=<azure-openai-endpoint>
5656
AZURE_OPENAI_API_KEY=<key>
57-
## it is recommended to use gpt-4o-mini, o1-mini, and o1-preview deployments
57+
## 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.
5858
AZURE_GPT4OMINI_DEPLOYMENT_NAME=<gpt-4o-mini-deployment-name>
5959
AZURE_O1MINI_DEPLOYMENT_NAME=[<o1-mini-deployment-name>]
60-
AZURE_O1_DEPLOYMENT_NAME=[<o1-preview-deployment-name>]
61-
## gpt-4o is optional
62-
AZURE_GPT4O_DEPLOYMENT_NAME=<gpt-4o-deployment-name, modelVersion: 2024-08-06>
63-
## gpt-4-turbo is optional: it’s expensive but provides better reviews than gpt-4o
64-
AZURE_GPT4TURBO_DEPLOYMENT_NAME=[<gpt-4-turbo-deployment-name, modelVersion: turbo-2024-04-09>]
60+
AZURE_O1_DEPLOYMENT_NAME=[<o1-deployment-name>]
6561

6662
# if using AWS Bedrock
6763
AWS_ACCESS_KEY_ID=<aws-access-key>
@@ -108,7 +104,7 @@ docker pull us-docker.pkg.dev/coderabbitprod/self-hosted/coderabbit-agent:latest
108104

109105
### Verify the image is up
110106

111-
You can query `/health` endpoint to verify that the coderabbit-agent service is up and running.
107+
You can query `/health` endpoint to verify that the `coderabbit-agent` service is up and running.
112108

113109
```bash
114110
curl 127.0.0.1:8080/health

Diff for: ‎docs/self-hosted/github.md

+3-7
Original file line numberDiff line numberDiff line change
@@ -65,14 +65,10 @@ LLM_PROVIDER=azure-openai
6565
LLM_TIMEOUT=360000
6666
AZURE_OPENAI_ENDPOINT=<azure-openai-endpoint>
6767
AZURE_OPENAI_API_KEY=<key>
68-
## it is recommended to use gpt-4o-mini, o1-mini, and o1-preview deployments
68+
## 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.
6969
AZURE_GPT4OMINI_DEPLOYMENT_NAME=<gpt-4o-mini-deployment-name>
7070
AZURE_O1MINI_DEPLOYMENT_NAME=[<o1-mini-deployment-name>]
71-
AZURE_O1_DEPLOYMENT_NAME=[<o1-preview-deployment-name>]
72-
## gpt-4o is optional
73-
AZURE_GPT4O_DEPLOYMENT_NAME=<gpt-4o-deployment-name, modelVersion: 2024-08-06>
74-
## gpt-4-turbo is optional: it’s expensive but provides better reviews than gpt-4o
75-
AZURE_GPT4TURBO_DEPLOYMENT_NAME=[<gpt-4-turbo-deployment-name, modelVersion: turbo-2024-04-09>]
71+
AZURE_O1_DEPLOYMENT_NAME=[<o1-deployment-name>]
7672

7773
# if using AWS Bedrock
7874
AWS_ACCESS_KEY_ID=<aws-access-key>
@@ -145,7 +141,7 @@ docker run --env-file .env --publish 127.0.0.1:8080:8080 <docker-registry>/coder
145141

146142
### Verify the image is up
147143

148-
You can query `/health` endpoint to verify that the coderabbit-agent service is up and running.
144+
You can query `/health` endpoint to verify that the `coderabbit-agent` service is up and running.
149145

150146
```bash
151147
curl 127.0.0.1:8080/health

Diff for: ‎docs/self-hosted/gitlab.md

+4-8
Original file line numberDiff line numberDiff line change
@@ -42,9 +42,9 @@ Consult official CodeRabbitAI documentation for a detailed [guide](https://docs.
4242
- Issues events
4343
- Merge request events
4444

45-
## Prepare an `.env` file
45+
## Prepare a `.env` file
4646

47-
Create an `.env` file with the following content:
47+
Create a `.env` file with the following content:
4848

4949
```bash
5050
# if using OpenAI
@@ -60,14 +60,10 @@ LLM_PROVIDER=azure-openai
6060
LLM_TIMEOUT=360000
6161
AZURE_OPENAI_ENDPOINT=<azure-openai-endpoint>
6262
AZURE_OPENAI_API_KEY=<key>
63-
## it is recommended to use gpt-4o-mini, o1-mini, and o1-preview deployments
63+
## 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.
6464
AZURE_GPT4OMINI_DEPLOYMENT_NAME=<gpt-4o-mini-deployment-name>
6565
AZURE_O1MINI_DEPLOYMENT_NAME=[<o1-mini-deployment-name>]
66-
AZURE_O1_DEPLOYMENT_NAME=[<o1-preview-deployment-name>]
67-
## gpt-4o is optional
68-
AZURE_GPT4O_DEPLOYMENT_NAME=<gpt-4o-deployment-name, modelVersion: 2024-08-06>
69-
## gpt-4-turbo is optional: it’s expensive but provides better reviews than gpt-4o
70-
AZURE_GPT4TURBO_DEPLOYMENT_NAME=[<gpt-4-turbo-deployment-name, modelVersion: turbo-2024-04-09>]
66+
AZURE_O1_DEPLOYMENT_NAME=[<o1-deployment-name>]
7167

7268
# if using AWS Bedrock
7369
AWS_ACCESS_KEY_ID=<aws-access-key>

0 commit comments

Comments
 (0)