Vercel

Vercel aggregates models from multiple providers with enhanced features like rate limiting and failover. Access 86 models through NotPixel’s model router.

Learn more in the Vercel AI SDK documentation .

Usage


import Ads from 'notpixel';
 
const ads = new Ads({
  publisherId: 'pub_xxx',
  model: 'vercel/openai/gpt-5.2',
  input: 'How do I deploy a Next.js app?',
});
 
const response = await ads.offer();
console.log(response.text);

Configuration

.env


VERCEL_API_KEY=your-gateway-key
 
# Or use provider API keys directly
OPENAI_API_KEY=sk-...
ANTHROPIC_API_KEY=sk-ant-...

Available Models

Model
`vercel/alibaba/qwen3-coder-plus`
`vercel/alibaba/qwen3-max`
`vercel/alibaba/qwen3-next-80b-a3b-instruct`
`vercel/alibaba/qwen3-next-80b-a3b-thinking`
`vercel/alibaba/qwen3-vl-instruct`
`vercel/alibaba/qwen3-vl-thinking`
`vercel/amazon/nova-lite`
`vercel/amazon/nova-micro`
`vercel/amazon/nova-pro`
`vercel/anthropic/claude-3-haiku`
`vercel/anthropic/claude-opus-4.5`
`vercel/anthropic/claude-3.5-haiku`
`vercel/anthropic/claude-3.5-sonnet`
`vercel/anthropic/claude-3.7-sonnet`
`vercel/anthropic/claude-4-1-opus`
`vercel/anthropic/claude-4-opus`
`vercel/anthropic/claude-4-sonnet`
`vercel/anthropic/claude-4.5-sonnet`
`vercel/anthropic/claude-haiku-4.5`
`vercel/anthropic/claude-opus-4.5`
`vercel/deepseek/deepseek-r1`
`vercel/deepseek/deepseek-r1-distill-llama-70b`
`vercel/deepseek/deepseek-v3.1-terminus`
`vercel/deepseek/deepseek-v3.2-exp`
`vercel/deepseek/deepseek-v3.2-exp-thinking`
`vercel/google/gemini-2.0-flash`
`vercel/google/gemini-2.0-flash-lite`
`vercel/google/gemini-2.5-flash`
`vercel/google/gemini-2.5-flash-lite`
`vercel/google/gemini-2.5-flash-lite-preview-09-2025`
`vercel/google/gemini-2.5-flash-preview-09-2025`
`vercel/google/gemini-2.5-pro`
`vercel/google/gemini-3-pro-preview`
`vercel/meta/llama-3.3-70b`
`vercel/meta/llama-4-maverick`
`vercel/meta/llama-4-scout`
`vercel/minimax/minimax-m2`
`vercel/mistral/codestral`
`vercel/mistral/magistral-medium`
`vercel/mistral/magistral-small`
`vercel/mistral/ministral-3b`
`vercel/mistral/ministral-8b`
`vercel/mistral/mistral-large`
`vercel/mistral/mistral-small`
`vercel/mistral/mixtral-8x22b-instruct`
`vercel/mistral/pixtral-12b`
`vercel/mistral/pixtral-large`
`vercel/moonshotai/kimi-k2`
`vercel/morph/morph-v3-fast`
`vercel/morph/morph-v3-large`
`vercel/openai/gpt-4-turbo`
`vercel/openai/gpt-4.1`
`vercel/openai/gpt-4.1-mini`
`vercel/openai/gpt-4.1-nano`
`vercel/openai/gpt-5.2`
`vercel/openai/gpt-5.2-mini`
`vercel/openai/gpt-5`
`vercel/openai/gpt-5-codex`
`vercel/openai/gpt-5-mini`
`vercel/openai/gpt-5-nano`
`vercel/openai/gpt-oss-120b`
`vercel/openai/gpt-oss-20b`
`vercel/openai/o1`
`vercel/openai/o3`
`vercel/openai/o3-mini`
`vercel/openai/o4-mini`
`vercel/perplexity/sonar`
`vercel/perplexity/sonar-pro`
`vercel/perplexity/sonar-reasoning`
`vercel/perplexity/sonar-reasoning-pro`
`vercel/vercel/v0-1.0-md`
`vercel/vercel/v0-1.5-md`
`vercel/xai/grok-2`
`vercel/xai/grok-2-vision`
`vercel/xai/grok-3`
`vercel/xai/grok-3-fast`
`vercel/xai/grok-3-mini`
`vercel/xai/grok-3-mini-fast`
`vercel/xai/grok-4`
`vercel/xai/grok-4-fast`
`vercel/xai/grok-4-fast-non-reasoning`
`vercel/xai/grok-code-fast-1`
`vercel/zai/glm-4.5`
`vercel/zai/glm-4.5-air`
`vercel/zai/glm-4.5v`
`vercel/zai/glm-4.6`

Model Format

Use the format vercel/{provider}/{model}:


// OpenAI via Vercel
model: 'vercel/openai/gpt-5.2'
 
// Anthropic via Vercel
model: 'vercel/anthropic/claude-3.5-sonnet'
 
// Google via Vercel
model: 'vercel/google/gemini-2.5-pro'
 
// Meta via Vercel
model: 'vercel/meta/llama-3.3-70b'

Why Use Vercel Gateway?

Unified API: Single interface for multiple providers
Rate limiting: Built-in rate limit management
Failover: Automatic failover between providers
Edge optimized: Low latency from Vercel’s edge network

Environment Variable

Variable	Description
`VERCEL_API_KEY`	Your Vercel gateway API key