Skip to content

Commit 25710d5

Browse files
Initial documentation for chat completions
1 parent 2a091c4 commit 25710d5

File tree

1 file changed

+84
-0
lines changed

1 file changed

+84
-0
lines changed
Lines changed: 84 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,84 @@
1+
---
2+
title: Chat Completion Endpoint
3+
pcx_content_type: configuration
4+
sidebar:
5+
order: 5
6+
---
7+
8+
Cloudflare's AI Gateway offers an OpenAI-compatible `/chat/completions` endpoint, enabling integration with multiple AI providers using a single URL. This feature simplifies the integration process, allowing for seamless switching between different models without significant code modifications.
9+
10+
## Key benefits
11+
12+
- Standardization: Provides a unified format compatible with OpenAI's schema, reducing the need for code refactoring.
13+
- Ease of Development: Switch between different models and providers quickly using a consistent API structure.
14+
15+
## Endpoint URL
16+
17+
```txt
18+
https://gateway.ai.cloudflare.com/v1/{account_id}/{gateway_id}/compat/chat/completions
19+
```
20+
21+
Replace `{account_id}` and `{gateway_id}` with your Cloudflare account and gateway IDs.
22+
23+
## Using the Unified Interface
24+
25+
Switch providers by changing the `model` and `apiKey` parameters.
26+
27+
## Model Parameter Format
28+
29+
Specify the model using `{provider}/{model}` format. For example:
30+
31+
- `openai/gpt-4o-mini`
32+
- `google-ai-studio/gemini-2.0-flash`
33+
- `anthropic/claude-3-haiku`
34+
35+
## Example with OpenAI JavaScript SDK
36+
37+
```js
38+
import OpenAI from "openai";
39+
const client = new OpenAI({
40+
apiKey: "YOUR_PROVIDER_API_KEY", // Provider API key
41+
baseURL:
42+
"https://gateway.ai.cloudflare.com/v1/{account_id}/{gateway_id}/compat/chat/completions",
43+
});
44+
45+
const response = await client.chat.completions.create({
46+
model: "google-ai-studio/gemini-2.0-flash",
47+
messages: [{ role: "user", content: "What is Cloudflare?" }],
48+
});
49+
50+
console.log(response.choices[0].message.content);
51+
```
52+
53+
## `curl` Example
54+
55+
```bash
56+
curl -X POST https://gateway.ai.cloudflare.com/v1/{account_id}/{gateway_id}/compat/chat/completions \
57+
--header 'Authorization: Bearer {openai_token}' \
58+
--header 'Content-Type: application/json' \
59+
--data '{
60+
"model": "google-ai-studio/gemini-2.0-flash",
61+
"messages": [
62+
{
63+
"role": "user",
64+
"content": "What is Cloudflare?"
65+
}
66+
]
67+
}'
68+
```
69+
70+
## Supported Providers
71+
72+
The OpenAI-compatible endpoint supports models from the following providers:
73+
74+
- [Anthropic](/ai-gateway/providers/anthropic/)
75+
- [OpenAI](/ai-gateway/providers/openai/)
76+
- [Groq](/ai-gateway/providers/groq/)
77+
- [Mistral](/ai-gateway/providers/mistral/)
78+
- [Cohere](/ai-gateway/providers/cohere/)
79+
- [Perplexity](/ai-gateway/providers/perplexity/)
80+
- [Workers AI](/ai-gateway/providers/workersai/)
81+
- [Google-AI-Studio](/ai-gateway/providers/google-ai-studio/)
82+
- [Grok](/ai-gateway/providers/grok/)
83+
- [DeepSeek](/ai-gateway/providers/deepseek/)
84+
- [Cerebras](/ai-gateway/providers/cerebras/)

0 commit comments

Comments
 (0)