Skip to main content

Documentation Index

Fetch the complete documentation index at: https://narev.ai/docs/llms.txt

Use this file to discover all available pages before exploring further.

​
Endpoint

POST /api/applications/{application_id}/v1/chat/completions

​
Authentication

Include your Narev API key in the Authorization header: 
Authorization: Bearer YOUR_API_KEY
You can generate API keys in the Narev Cloud dashboard under Settings → API Keys.

​
Setup

from openai import OpenAI

client = OpenAI(
    api_key="YOUR_API_KEY",
    base_url="https://narev.ai/api/applications/{application_id}/v1"
)

​
Request parameters

​
Required

​
messages
array
required
Array of message objects, each with a role (system, user, or assistant) and content string.

​
Optional

Parameter support varies by model. Check your model’s documentation to confirm which parameters it accepts.
​
model
string
Model identifier with gateway prefix (for example, openai:gpt-4). If omitted, Narev uses the A/B test’s production variant.
​
temperature
number
Sampling temperature between 0 and 2. Higher values produce more random output.
​
top_p
number
Nucleus sampling parameter between 0 and 1. Lower values make output more focused.
​
top_k
integer
Limits sampling to the K most likely next tokens.
​
max_tokens
integer
Maximum number of tokens to generate in the response.
​
frequency_penalty
number
Penalizes tokens based on their frequency in the text so far. Range: -2.0 to 2.0.
​
presence_penalty
number
Penalizes tokens that have already appeared in the text so far. Range: -2.0 to 2.0.
​
repetition_penalty
number
Penalizes repeated tokens. Typical range: 0.0 to 2.0.
​
min_p
number
Minimum probability threshold for token selection. Range: 0 to 1.
​
seed
integer
Random seed for deterministic generation.
​
logprobs
boolean
When true, returns log probabilities for each output token.
​
top_logprobs
integer
Number of top log probabilities to return. Range: 0 to 20. Requires logprobs: true.
​
response_format
object
Controls the format of the response. Pass {"type": "json_object"} to enable JSON mode.
​
stop
string | array
Up to four sequences at which the API stops generating further tokens.
​
stream
boolean
default:"false"
When true, Narev streams the response as server-sent events (SSE).
​
metadata
object
Custom metadata for tracking and automatic quality evaluation.
FieldTypeDescription
expected_outputstringExpected response text for automatic quality scoring

​
Model identifiers

Models use a {gateway}:{model_name} format:
GatewayExample
openaiopenai:gpt-4
anthropicanthropic:claude-3-opus-20240229
openrouteropenrouter:meta-llama/llama-3.1-70b-instruct
vertexvertex:gemini-pro
bedrockbedrock:amazon.titan-text-express-v1
portkeyportkey:gpt-4
heliconehelicone:gpt-4

​
Request examples

​
Basic request

response = client.chat.completions.create(
    model="openai:gpt-4",
    messages=[
        {"role": "user", "content": "What is the capital of France?"}
    ]
)

​
With system prompt

response = client.chat.completions.create(
    model="openai:gpt-4",
    messages=[
        {"role": "system", "content": "You are a helpful geography expert."},
        {"role": "user", "content": "What is the capital of France?"}
    ]
)

​
With generation parameters

response = client.chat.completions.create(
    model="openai:gpt-4",
    messages=[
        {"role": "user", "content": "Write a creative story."}
    ],
    temperature=0.9,
    max_tokens=500,
    top_p=0.95
)

​
With JSON response format

response = client.chat.completions.create(
    model="openai:gpt-4",
    messages=[
        {"role": "user", "content": "Return user data as JSON."}
    ],
    response_format={"type": "json_object"}
)

​
Streaming

stream = client.chat.completions.create(
    model="openai:gpt-4",
    messages=[
        {"role": "user", "content": "Tell me a story."}
    ],
    stream=True
)

for chunk in stream:
    if chunk.choices[0].delta.content:
        print(chunk.choices[0].delta.content, end="")

​
With quality evaluation

response = client.chat.completions.create(
    model="openai:gpt-4",
    messages=[
        {"role": "user", "content": "What is the capital of France?"}
    ],
    extra_body={
        "metadata": {
            "expected_output": "Paris is the capital of France."
        }
    }
)

​
Response format

​
Non-streaming

{
  "id": "chatcmpl-123",
  "object": "chat.completion",
  "created": 1677652288,
  "model": "openai:gpt-4",
  "choices": [
    {
      "index": 0,
      "message": {
        "role": "assistant",
        "content": "Paris is the capital of France."
      },
      "finish_reason": "stop"
    }
  ],
  "usage": {
    "prompt_tokens": 13,
    "completion_tokens": 7,
    "total_tokens": 20
  }
}

​
Streaming

Narev sends each token as a server-sent event (SSE) with a data: prefix: 
data: {"id":"chatcmpl-123","object":"chat.completion.chunk","created":1677652288,"model":"openai:gpt-4","choices":[{"index":0,"delta":{"content":"Paris"},"finish_reason":null}]}

data: {"id":"chatcmpl-123","object":"chat.completion.chunk","created":1677652288,"model":"openai:gpt-4","choices":[{"index":0,"delta":{"content":" is"},"finish_reason":null}]}

data: [DONE]

​
Error responses

All errors return a JSON object with an error field: 
{
  "error": {
    "message": "Error description",
    "code": "error_code"
  }
}
StatusCodeDescription
400bad_requestInvalid request format or parameters
400model_requiredModel is required when no production variant is set
401invalid_api_keyInvalid or missing API key
402insufficient_creditsInsufficient credits to complete the request
404application_not_foundA/B test ID not found
500internal_errorInternal server error