Neurometric API Documentation
Drop-in replacement for OpenAI with built-in analytics and cost optimization
Quick Start
Replace your OpenAI endpoint with Neurometric in just one line. All your existing code works exactly the same.
2. Update your code
Just change the base_url. Everything else stays the same.
from openai import OpenAI
client = OpenAI(
api_key=os.environ.get("NEUROMETRIC_API_KEY"),
base_url="https://api.neurometric.ai/v1"
)
response = client.chat.completions.create(
model="gpt-4o",
messages=[
{"role": "user", "content": "Hello!"}
]
)
print(response.choices[0].message.content)import OpenAI from 'openai';
const client = new OpenAI({
apiKey: process.env.NEUROMETRIC_API_KEY,
baseURL: 'https://api.neurometric.ai/v1',
});
const response = await client.chat.completions.create({
model: 'gpt-4o',
messages: [
{ role: 'user', content: 'Hello!' }
],
});
console.log(response.choices[0].message.content);curl -X POST https://api.neurometric.ai/v1/chat/completions \
-H "Authorization: Bearer $NEUROMETRIC_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "gpt-4o",
"messages": [{"role": "user", "content": "Hello!"}]
}'Analytics & Observability
Zero Configuration Required
Analytics are automatic. Every request through Neurometric is tracked, analyzed, and available in your dashboard without any extra code or setup.
Usage Tracking
Monitor all API calls, models used, tokens consumed, and response times in real-time.
Cost Analysis
Automatic cost breakdown by model, task type, and time period. Identify optimization opportunities.
Performance Metrics
Track latency, throughput, error rates, and success metrics across all your LLM requests.
Smart Recommendations
Get AI-powered suggestions to reduce costs while maintaining quality based on your usage patterns.
View Your Analytics Dashboard
Access detailed analytics, usage reports, and optimization recommendations.
Open DashboardAPI Reference
Base URL
Authentication
All API requests require authentication using a Bearer token in the Authorization header.
Authorization: Bearer YOUR_API_KEYEndpoint
/chat/completionsCreate a chat completion. Fully compatible with OpenAI's chat completions API format.
Supported Models
Use any model ID from OpenAI, Anthropic, Google, or other providers. Neurometric automatically routes to the correct provider.
OpenAI
gpt-4o, gpt-4-turbo, gpt-3.5-turbo
Anthropic
claude-3-5-sonnet, claude-3-opus
gemini-pro, gemini-flash
Meta
llama-3-70b, llama-3-8b
Mistral
mistral-large, mistral-7b
...and more
100+ models supported
Request Format
Required
modelstringID of the model to use. See supported models above.
Example: "gpt-4o", "claude-3-5-sonnet"
messagesarrayArray of message objects with role and content.
Roles: "system", "user", "assistant"
Optional
temperaturenumberSampling temperature between 0 and 2. Higher values = more random.
Default: 1
max_tokensintegerMaximum number of tokens to generate in the response.
Model-dependent default
top_pnumberNucleus sampling: consider tokens with top_p probability mass.
Default: 1
stopstring | arraySequences where the API will stop generating tokens.
Up to 4 sequences
Example Request
{
"model": "gpt-4o",
"messages": [
{
"role": "system",
"content": "You are a helpful assistant."
},
{
"role": "user",
"content": "What is the capital of France?"
}
],
"temperature": 0.7,
"max_tokens": 150
}Response Format
idstringUnique identifier for the completion.
modelstringThe model that generated the response.
choicesarrayArray of completion choices. Each contains:
message - The generated message object
finish_reason - Why generation stopped ("stop", "length", etc.)
index - Index of this choice
usageobjectToken usage statistics:
prompt_tokens - Tokens in the prompt
completion_tokens - Tokens in the response
total_tokens - Total tokens used
Example Response
{
"id": "chatcmpl-abc123",
"object": "chat.completion",
"created": 1706745600,
"model": "gpt-4o",
"choices": [
{
"index": 0,
"message": {
"role": "assistant",
"content": "The capital of France is Paris."
},
"finish_reason": "stop"
}
],
"usage": {
"prompt_tokens": 25,
"completion_tokens": 8,
"total_tokens": 33
}
}Error Handling
Errors follow a standard format with HTTP status codes and descriptive messages:
{ "error": { "message": "...", "type": "...", "code": "..." } }401 - Invalid API key
429 - Rate limit exceeded
500 - Server error
TypeScript Types
If you're using TypeScript, here are the type definitions for request and response objects.
Type Definitions
// Message in a conversation
interface Message {
role: 'system' | 'user' | 'assistant';
content: string;
}
// Request body for /chat/completions
interface ChatCompletionRequest {
model: string;
messages: Message[];
temperature?: number; // 0-2, default 1
max_tokens?: number; // Model-dependent
top_p?: number; // 0-1, default 1
stop?: string | string[]; // Up to 4 sequences
}
// Response from /chat/completions
interface ChatCompletionResponse {
id: string;
object: 'chat.completion';
created: number;
model: string;
choices: {
index: number;
message: Message;
finish_reason: 'stop' | 'length' | 'content_filter';
}[];
usage: {
prompt_tokens: number;
completion_tokens: number;
total_tokens: number;
};
}Migration from OpenAI
Switch in 3 Steps
- 1
Get your Neurometric API key
Create an account and generate an API key
- 2
Update your base URL
Change
https://api.openai.com/v1tohttps://api.neurometric.ai/v1 - 3
Replace your API key
Use your Neurometric API key instead of your OpenAI key
Full Compatibility
No other code changes required. Neurometric uses the same request/response format as OpenAI, so your existing code works without modification. You can even continue using the official OpenAI SDK.
Support
Need help? We're here for you.