Netwrck

LLM Models API

OpenAI-compatible chat completions with 15+ models and automatic fallbacks

Overview

Access multiple AI models through a single, OpenAI-compatible API endpoint. Use your existing OpenAI SDK or HTTP client - just change the base URL and API key.

Automatic Fallback & Refusal Handling

If a model is down, rate-limited, or refuses your request, we automatically retry with alternative models to maximize request success. You only pay for the model that generates the response.

Quick Start

Using curl

curl https://netwrck.com/v1/chat/completions \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "thedrummer/skyfall-36b-v2",
    "messages": [{"role": "user", "content": "Hello!"}],
    "max_tokens": 1000
  }'

Using Python (OpenAI SDK)

from openai import OpenAI

client = OpenAI(
    base_url="https://netwrck.com/v1",
    api_key="YOUR_API_KEY"
)

response = client.chat.completions.create(
    model="thedrummer/skyfall-36b-v2",
    messages=[{"role": "user", "content": "Hello!"}],
    max_tokens=1000
)

print(response.choices[0].message.content)

Using JavaScript

import OpenAI from 'openai';

const client = new OpenAI({
  baseURL: 'https://netwrck.com/v1',
  apiKey: 'YOUR_API_KEY'
});

const response = await client.chat.completions.create({
  model: 'thedrummer/skyfall-36b-v2',
  messages: [{ role: 'user', content: 'Hello!' }],
  max_tokens: 1000
});

console.log(response.choices[0].message.content);

API Endpoints

POST /v1/chat/completions

Create a chat completion (OpenAI-compatible)

Request Body

{
  "model": "thedrummer/skyfall-36b-v2",
  "messages": [
    {"role": "system", "content": "You are a helpful assistant."},
    {"role": "user", "content": "Hello!"}
  ],
  "max_tokens": 1000,
  "temperature": 0.7
}

Response

{
  "id": "chatcmpl-abc123",
  "object": "chat.completion",
  "created": 1703001234,
  "model": "thedrummer/skyfall-36b-v2",
  "choices": [{
    "index": 0,
    "message": {
      "role": "assistant",
      "content": "Hello! How can I help you today?"
    },
    "finish_reason": "stop"
  }],
  "usage": {
    "prompt_tokens": 15,
    "completion_tokens": 10,
    "total_tokens": 25
  }
}

GET /v1/models

List available models with pricing

GET /api/credits/balance

Check your credit balance

Character Chat API

Use /api/chatpred_or to chat with an existing Netwrck character by exact name or URL name. The endpoint loads the character description and greeting when context is omitted.

curl -X POST https://netwrck.com/api/chatpred_or \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "character": "Sherlock Holmes",
    "query": "What do you notice?",
    "model_name": "deepseek/deepseek-v4-flash",
    "use_usage_credits": true
  }'

Use context for prior messages or custom instructions. Supplying it overrides the character's automatic description context.

Available Models

All prices are per 1 million tokens.

Model	Provider	Input	Output	Tags
`inclusionai/ling-3.0-flash`	InclusionAI	$0.000	$0.000	generalfastfree
`thedrummer/skyfall-36b-v2`	TheoDrummer	$0.660	$0.960	creativeroleplayfreeenhanced
`gryphe/mythomax-l2-13b`	Gryphe	$0.072	$0.072	generalreliablefree
`openai/gpt-5.6-luna`	OpenAI	$0.240	$1.440	premiumfastopenai
`openai/gpt-4.1-mini`	OpenAI	$0.180	$0.720	premiumenhancedopenai
`openai/gpt-5.2`	OpenAI	$2.400	$21.600	premiumreasoninglatestopenai
`minimax/minimax-m2.5`	MiniMax	$0.360	$1.320	premiumcodingreasoninglatest
`anthropic/claude-sonnet-5`	Anthropic	$3.600	$18.000	premiumcodingreasoninganthropic
`anthropic/claude-fable-5`	Anthropic	$12.000	$60.000	premiumcodingreasoninganthropiclatest
`google/gemini-pro`	Google	$0.600	$1.800	premiummultimodalgoogle
`google/gemini-flash`	Google	$0.090	$0.360	premiumfastgoogle
`google/gemini-2.5-flash`	Google	$0.180	$0.720	premiumfastgooglelatest
`deepseek/deepseek-r1`	DeepSeek	$0.660	$2.628	codingreasoningthinkingpremium
`deepseek/deepseek-chat`	DeepSeek	$0.168	$0.336	codingdevelopmentpremium
`deepseek/deepseek-v4-flash`	DeepSeek	$0.168	$0.336	codinglatestpremiumfast

Automatic Fallbacks

Our API automatically handles model failures and refusals to maximize reliability:

Model downtime: If a model is unavailable, we automatically retry with alternative models
Rate limiting: If you hit provider limits, requests are routed to fallback models
Content refusals: If a model refuses a request (e.g., "I cannot help with that"), we automatically retry with more permissive models

How Fallbacks Work

The response's model field shows which model actually generated the response. You only pay for the model used, not the original request. Fallback order prioritizes cheaper, more reliable models to keep costs low.

Fallback requests currently use gryphe/mythomax-l2-13b.

Authentication

All API requests require authentication via Bearer token in the Authorization header:

Authorization: Bearer YOUR_API_KEY

Get your API key from your account settings.

Billing

You're billed per token based on the model used:

Input tokens: Charged for your prompt/messages
Output tokens: Charged for the model's response
Token estimation: ~1.3 tokens per word (English)

Credits are deducted in USD. Check your balance anytime via the /api/credits/balance endpoint or the credit display in the header.

Example Cost Calculation

For a request with 500 input tokens and 1000 output tokens using Skyfall 36B:

Input:  500 tokens  × ($0.66 / 1M) = $0.00033
Output: 1000 tokens × ($0.96 / 1M) = $0.00096
Total:  $0.00129 per request

Error Handling

Errors follow the OpenAI error format:

{
  "error": {
    "message": "Insufficient credits: have $0.50, need ~$1.00",
    "type": "insufficient_quota",
    "code": 402
  }
}

Code	Type	Description
400	invalid_request_error	Missing or invalid parameters
401	invalid_api_key	Invalid or missing API key
402	insufficient_quota	Not enough credits
500	api_error	Server error (all models failed)