Models
NotPixel SDK supports multiple LLM providers and gateways, allowing you to choose the best model for your use case.
Model Providers
Direct access to individual AI providers. Configure your API key and start using their models.
OpenAIAnthropicGooglexAIDeepSeekGroqMistralPerplexityTogether AIGitHub ModelsNVIDIAAlibabaMinimaxLlamaLM Studio
Model Gateways
Unified APIs that provide access to multiple providers through a single API key.
Quick Start
Configure your API key:
.env
OPENAI_API_KEY=sk-...Use the model:
import Ads from 'notpixel';
const ads = new Ads({
publisherId: 'pub_xxx',
model: 'openai/gpt-5.2',
input: 'Your user query here',
});
const response = await ads.offer();
console.log(response.text);The SDK automatically detects API keys from your environment.
Model Format
All models use the format {provider}/{model-name}:
| Format | Example |
|---|---|
openai/{model} | openai/gpt-5.2, openai/gpt-5.2-mini |
anthropic/{model} | anthropic/claude-opus-4.5 |
google/{model} | google/gemini-3-pro |
xai/{model} | xai/grok-2 |
deepseek/{model} | deepseek/deepseek-chat |
groq/{model} | groq/llama-3.1-70b-versatile |
mistral/{model} | mistral/mistral-large-latest |
perplexity/{model} | perplexity/sonar |
together/{model} | together/deepseek-ai/DeepSeek-R1 |
github/{model} | github/gpt-4o |
nvidia/{model} | nvidia/meta/llama-3.1-70b-instruct |
alibaba/{model} | alibaba/qwen-max |
minimax/{model} | minimax/MiniMax-M2 |
llama/{model} | llama/llama-3.3-70b-instruct |
lmstudio/{model} | lmstudio/local-model |
openrouter/{provider}/{model} | openrouter/anthropic/claude-opus-4.5 |
vercel/{provider}/{model} | vercel/openai/gpt-5.2 |
netlify/{provider}/{model} | netlify/anthropic/claude-3-5-haiku |
Choosing a Model
| Use Case | Recommended Model |
|---|---|
| General purpose | openai/gpt-5.2 |
| Fast & cheap | openai/gpt-5.2-mini, groq/llama-3.1-8b |
| Complex analysis | anthropic/claude-opus-4.5 |
| Long documents | google/gemini-3-pro |
| Real-time info | xai/grok-2, perplexity/sonar |
| Code generation | deepseek/deepseek-coder |
| Local/private | lmstudio/local-model |