/ai (AI chat proxy)

Proxy to an OpenAI-compatible API via GitHub Models.

Methods

GET, POST

Query parameter

Parameter	Type	Required	Description
`query`	string	required if no `messages` provided	Simple prompt text used to build messages.

JSON request body (pass-through)

Content-Type: application/json

Field	Type	Required	Description
`messages`	array	required if no `query` provided	Chat history array passed to the model.
`model`	string	no	Overrides the default model.
`temperature`	number	no	Sampling temperature passed through to the model.
`max_tokens`	integer	no	Maximum tokens for the response.

Response

{
  "answer": "...generated text..."
}

Errors

400 Missing query and messages.
500 Upstream API error.

Example

curl -G "https://sodeom.com/ai" --data-urlencode "query=Say hi"

curl -X POST "https://sodeom.com/ai" \
  -H "Content-Type: application/json" \
  -d '{"model":"gpt-4o-mini","messages":[{"role":"user","content":"Say hi"}],"temperature":0.7}'

Notes

Requires GITHUB_TOKEN on the server.
Default model: gpt-4o-mini unless overridden in server config.

OpenAI SDK-compatible API

Use any OpenAI SDK or tool by pointing it at https://sodeom.com/v1 as the base URL.

Quick start

from openai import OpenAI

client = OpenAI(
    base_url="https://sodeom.com/v1",
    api_key="any",          # no key required, pass anything
)

response = client.chat.completions.create(
    model="gpt-4o-mini",
    messages=[{"role": "user", "content": "Hello!"}],
)
print(response.choices[0].message.content)

Streaming

stream = client.chat.completions.create(
    model="gpt-4o-mini",
    messages=[{"role": "user", "content": "Count to 5"}],
    stream=True,
)
for chunk in stream:
    print(chunk.choices[0].delta.content or "", end="", flush=True)

Endpoints

Method	Path	Description
`GET`	`/v1/models`	List available models.
`GET`	`/v1/models/:id`	Retrieve a single model.
`POST`	`/v1/chat/completions`	Create a chat completion (streaming supported).

Available models

gpt-4o-mini (default)
gpt-4o
o1-mini
Meta-Llama-3.1-8B-Instruct
Meta-Llama-3.1-70B-Instruct
Mistral-small
Phi-3.5-mini-instruct

Supported parameters

Parameter	Type	Description
`model`	string	Model ID (default: gpt-4o-mini).
`messages`	array	Required. Chat history.
`stream`	boolean	Enable SSE streaming.
`temperature`	number	Sampling temperature.
`max_tokens`	integer	Max tokens (default: 1024).
`top_p`	number	Nucleus sampling.
`stop`	string/array	Stop sequences.
`frequency_penalty`	number	Frequency penalty.
`presence_penalty`	number	Presence penalty.
`seed`	integer	Deterministic seed.
`n`	integer	Number of completions.

curl examples

# List models
curl https://sodeom.com/v1/models

# Chat completion
curl https://sodeom.com/v1/chat/completions \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-4o-mini",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

# Streaming
curl https://sodeom.com/v1/chat/completions \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-4o-mini",
    "messages": [{"role": "user", "content": "Count to 5"}],
    "stream": true
  }'