Chat Completions

The Chat Completions endpoint is the primary way to interact with ORVEXA models. It is fully compatible with the OpenAI Chat Completions API.

POST

/v1/chat/completionsOpenAI-compatible chat endpoint

Request Parameters

Parameter	Type	Required	Description
`model`	string	Required	The ID of the model to use (e.g., deepseek-chat).
`messages`	array	Required	An array of message objects containing the conversation history. Each object has a role (system, user, or assistant) and content.
`temperature`	number	Optional	Sampling temperature between 0 and 2. Higher values make output more random; lower values make it more focused. Default is 1.
`max_tokens`	integer	Optional	The maximum number of tokens to generate in the response.
`stream`	boolean	Optional	If set to true, partial message deltas are sent as server-sent events (SSE) as they become available.
`top_p`	number	Optional	Nucleus sampling parameter. The model considers results of tokens with top_p probability mass. Default is 1.
`frequency_penalty`	number	Optional	Number between -2.0 and 2.0. Positive values penalize new tokens based on their frequency in the text so far.
`presence_penalty`	number	Optional	Number between -2.0 and 2.0. Positive values penalize new tokens based on whether they appear in the text so far.
`stop`	string \| array	Optional	Up to 4 sequences where the API will stop generating further tokens.

Code Examples

bash

curl https://api.orvexaproject.com/v1/chat/completions \
  -H "Authorization: Bearer nx-sk-dcfab6de7407c3c5f74fb627" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "deepseek-chat",
    "messages": [
      {"role": "system", "content": "You are a helpful assistant."},
      {"role": "user", "content": "Write a haiku about coding."}
    ],
    "temperature": 0.7,
    "max_tokens": 150
  }'

Response Format

For non-streaming requests, the API returns a complete JSON object once the model finishes generating.

Non-streaming responsejson

{
  "id": "chatcmpl-abc123",
  "object": "chat.completion",
  "created": 1718000000,
  "model": "deepseek-chat",
  "choices": [
    {
      "index": 0,
      "message": {
        "role": "assistant",
        "content": "Keys click and tap bright,\nLogic flows through silent lines,\nBugs flee into night."
      },
      "finish_reason": "stop"
    }
  ],
  "usage": {
    "prompt_tokens": 20,
    "completion_tokens": 17,
    "total_tokens": 37
  }
}

Streaming Response

When stream is set to true, the response is sent as server-sent events (SSE). Each chunk contains a delta object with the incremental content. The stream ends with a [DONE] message.

Streaming chunkstext

data: {"id":"chatcmpl-abc123","object":"chat.completion.chunk","choices":[{"index":0,"delta":{"content":"Keys"},"finish_reason":null}]}

data: {"id":"chatcmpl-abc123","object":"chat.completion.chunk","choices":[{"index":0,"delta":{"content":" click"},"finish_reason":null}]}

data: {"id":"chatcmpl-abc123","object":"chat.completion.chunk","choices":[{"index":0,"delta":{},"finish_reason":"stop"}]}

data: [DONE]

ORVEXA-Specific Headers

ORVEXA extends the OpenAI API with additional HTTP headers for cost control, caching, and multi-model fusion.

X-Orvexa-Max-CostSets a maximum cost limit in USD for the request. If the estimated cost exceeds this value, the request is rejected with a 402 error.

X-Response-CacheWhen set to true, identical requests within the cache TTL return cached responses, reducing latency and cost.

X-Prompt-CacheEnables automatic prompt caching. Repeated prefixes in messages are cached to reduce input token costs.

X-Orvexa-FusionEnables multi-model fusion mode. See the Fusion documentation for details.

Request with custom headersbash

curl https://api.orvexaproject.com/v1/chat/completions \
  -H "Authorization: Bearer nx-sk-dcfab6de7407c3c5f74fb627" \
  -H "Content-Type: application/json" \
  -H "X-Orvexa-Max-Cost: 0.05" \
  -H "X-Response-Cache: true" \
  -H "X-Prompt-Cache: true" \
  -d '{
    "model": "deepseek-chat",
    "messages": [{"role": "user", "content": "Hello"}]
  }'