Models API

Endpoints

Method	Path	Request	Response
GET	/v1/models	—	ModelList
POST	/v1/chat/completions	ChatCompletionRequest	ChatCompletion or an SSE stream

List models

GET /v1/models

curl https://api.dexto.ai/v1/models \
  -H "Authorization: Bearer $DEXTO_API_KEY"

Response field	JSON type	Required	Description
object	"list"	Yes	List discriminator
data	array<Model>	Yes	Models Dexto supports; your account may have different access methods
data[].id	string	Yes	Model identifier used in requests; replies name the model by this id
data[].object	"model"	Yes	Object discriminator
data[].name	string	Yes	Display name
data[].owned_by	string	Yes	The lab that makes the model, such as anthropic or openai
data[].context_window	integer	Yes	Most input tokens the model takes in one request
data[].max_output_tokens	integer|null	Yes	Most tokens the model writes in one reply; null when Dexto does not know it
data[].reasoning_efforts	array<string>	Yes	reasoning_effort values a request may send; empty when the model takes none
data[].created	integer	No	Unix timestamp in seconds

Create chat completion

POST /v1/chat/completions

The endpoint supports text, model-dependent image, PDF, and audio inputs, tool calls, reasoning, and streaming. Available modalities depend on the selected model.

Request field	JSON type	Required	Description
model	string	Yes	ID returned by GET /v1/models
reasoning_effort	string	No	One of the model’s reasoning_efforts
messages	array<Message>	Yes	Ordered conversation messages
messages[].role	string	Yes	system, user, assistant, or tool
messages[].content	string | array	Yes	Text or structured multimodal parts
stream	boolean	No	Return server-sent events
temperature	number	No	Sampling temperature
max_tokens	integer	No	Maximum generated tokens
tools	array<Tool>	No	OpenAI-compatible function tools

Request

curl https://api.dexto.ai/v1/chat/completions \
  -H "Authorization: Bearer $DEXTO_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "openai/gpt-5.6-luna",
    "messages": [{"role":"user","content":"Summarize this release note."}]
  }'

Response

{
  "id": "chatcmpl_01K4M6V7P8",
  "object": "chat.completion",
  "created": 1789142400,
  "model": "openai/gpt-5.6-luna",
  "choices": [{
    "index": 0,
    "message": {"role":"assistant","content":"The release expands the developer API."},
    "finish_reason": "stop"
  }],
  "usage": {"prompt_tokens":14,"completion_tokens":9,"total_tokens":23}
}

With stream: true, the response uses text/event-stream and returns OpenAI-compatible completion chunks.