Documentation

Unified AI API Gateway 2026: One API for Claude, GPT, Gemini, and 100+ Models

Stop managing multiple AI API keys. Access Claude, GPT-5, Gemini, DeepSeek, image generation, and video AI with one unified API. $2 free credits.

Managing multiple AI providers is painful. Different SDKs, separate billing, various rate limits. A unified AI API gateway solves this with one API key for everything.

The Multi-Provider Problem

Building with AI in 2026 means juggling:

ProviderModelsSDKBillingRate Limits
AnthropicClaudeanthropic-sdkPer orgComplex
OpenAIGPTopenai-sdkPer orgTiered
GoogleGeminigoogle-generativeaiPer projectDifferent
DeepSeekDeepSeekopenai-compatibleSeparateVariable
Model hosts100svariousPer predictionVariable

That's 5+ API keys, 5+ billing dashboards, 5+ sets of documentation.

The Solution: Unified AI Gateway

SkillBoss provides one API for all of them:

from openai import OpenAI

client = OpenAI(
    api_key="sk_your_single_key",
    base_url="https://api.skillboss.co/v1"
)

# Use ANY model with the same code
claude = client.chat.completions.create(model="claude-4-5-sonnet", messages=[...])
gpt = client.chat.completions.create(model="gpt-5", messages=[...])
gemini = client.chat.completions.create(model="gemini-2.5-pro", messages=[...])
deepseek = client.chat.completions.create(model="deepseek-r1", messages=[...])

What's Included

Language Models (40+)

Anthropic Claude

  • Claude 4.5 Opus (most intelligent)
  • Claude 4.5 Sonnet (best for coding)
  • Claude Haiku (fast, affordable)

OpenAI

  • GPT-5 (flagship)
  • GPT-4.1, o3-mini, o4-mini (reasoning)

Google

  • Gemini 2.5 Pro (1M context)
  • Gemini 2.5 Flash (ultra fast)

Open Source

  • DeepSeek R1, V3
  • Llama 3.1 405B, 70B
  • Mistral Large
  • Command R+

Image Generation (15+)

  • DALL-E 3
  • Flux Schnell, Dev
  • Stable Diffusion XL
  • Midjourney-style models

Video Generation (10+)

  • Google Veo 3.1
  • Text-to-video
  • Image-to-video
  • Video upscaling

Audio (15+)

  • Whisper (speech-to-text)
  • ElevenLabs (text-to-speech)
  • Voice cloning
  • Music generation

Data & Web (20+)

  • Firecrawl (web scraping)
  • Perplexity (AI search)
  • Google Maps
  • News, weather, stocks

Why Use a Unified Gateway?

1. One API Key

# Before: 5+ environment variables
ANTHROPIC_API_KEY=...
OPENAI_API_KEY=...
GOOGLE_API_KEY=...
DEEPSEEK_API_KEY=...
IMAGE_API_KEY=...

# After: 1 environment variable
SKILLBOSS_API_KEY=sk_your_key

2. One Billing Dashboard

See all usage in one place. No surprise bills from forgotten accounts.

3. Model Switching

Test different models without changing code:

# Just change the model parameter
model = "claude-4-5-sonnet"  # or "gpt-5", "gemini-2.5-pro"

4. Automatic Failover

If Claude is down, fallback to GPT automatically:

response = client.chat.completions.create(
    model="claude-4-5-sonnet",
    fallback_models=["gpt-5", "gemini-2.5-pro"],
    messages=[...]
)

5. Cost Optimization

Route to cheapest model for your use case:

# Use Claude for coding
code_response = client.chat.completions.create(model="claude-4-5-sonnet", ...)

# Use Gemini Flash for high-volume
bulk_response = client.chat.completions.create(model="gemini-2.5-flash", ...)

Integration Examples

Python

from openai import OpenAI

client = OpenAI(
    api_key="sk_your_key",
    base_url="https://api.skillboss.co/v1"
)

response = client.chat.completions.create(
    model="claude-4-5-sonnet",
    messages=[{"role": "user", "content": "Hello!"}]
)

JavaScript/TypeScript

import OpenAI from 'openai';

const client = new OpenAI({
  apiKey: 'sk_your_key',
  baseURL: 'https://api.skillboss.co/v1'
});

const response = await client.chat.completions.create({
  model: 'claude-4-5-sonnet',
  messages: [{ role: 'user', content: 'Hello!' }]
});

cURL

curl https://api.skillboss.co/v1/chat/completions \
  -H "Authorization: Bearer sk_your_key" \
  -d '{"model": "claude-4-5-sonnet", "messages": [...]}'

Works With All AI Tools

Claude Code

npm install -g skillboss-mcp-server

Cursor

Set base URL to https://api.skillboss.co/v1

Windsurf, Kiro, Codex

Same configurationβ€”OpenAI SDK compatible.

Pricing

Pay-as-you-go. Same rates as providers, sometimes lower.

ModelInput (per 1M)Output (per 1M)
Claude 4.5 Sonnet$3$15
GPT-5$10$30
Gemini 2.5 Flash$0.075$0.30
DeepSeek R1$0.55$2.19

$2 free credits to start. No subscription.

Compare: Unified AI Gateways

GatewayModelsFree TierAI Agent Support
SkillBoss100+$2 freeMCP Server
Other gateways50-100+LimitedNo
Enterprise gateways30-50+LimitedNo

Get Started

  1. Sign up: skillboss.co/console

Tip: For AI agents, use set up skillboss.co/skill.md to auto-configure everything.

  1. Get API key and $2 free credits
  2. Use 679+ endpoints with one API
# Install (optional, for Claude Code)
npm install -g skillboss-mcp-server

Related: