# Agent API

## One API. Frontier + open-source models. Zero friction.

Access every major AI model through a single unified endpoint. Drop-in OpenAI-compatible, so your existing code works instantly — just swap the base URL.

### Universal model access

Route requests to frontier and open-source models like Claude, GPT, Gemini, Grok, Llama, Mistral, DeepSeek, and Qwen through one API key. No separate accounts, no per-provider SDKs.

Automatic fallback routing ensures uptime even when individual providers experience outages. Smart load balancing distributes requests for optimal latency.

**api.blackbox.ai — model routing**

# Route to any model through one API key

→ "gpt-5.3-codex" routed to OpenAI  
→ "claude-sonnet" routed to Anthropic  
→ "gemini-3" routed to Google  
→ "grok-4" routed to xAI  
→ "llama-3" routed to Meta  
→ "mistral-large" routed to Mistral

### OpenAI-compatible endpoints

Use /v1/chat/completions, /v1/embeddings, and /v1/images/generations with the same request format you already know. Migrate in under 5 minutes.

Full streaming support, function calling, JSON mode, and vision inputs across all compatible models. Response format matches OpenAI spec exactly.

# Before (OpenAI SDK)

```plaintext
baseURL: "https://api.openai.com/v1"
apiKey: process.env.OPENAI_API_KEY
```

# After (Blackbox — same SDK)

```plaintext
baseURL: "https://api.blackbox.ai"
apiKey: process.env.BLACKBOX_API_KEY
```

✓ Same request format • same response shape

✓ Streaming, function calling, JSON mode

### Enterprise-grade infrastructure

99.9% uptime SLA, sub-200ms median latency, and usage-based billing with no per-seat costs. Scale from prototype to production without renegotiating contracts.

Enterprise-ready infrastructure with data residency options. Rate limits scale automatically with your plan. Real-time usage dashboards and cost alerts.

### FAQ

## Common Questions.

**IS THE API OPENAI-COMPATIBLE?**  
Yes. Use the same SDK and request format you already know — just swap the base URL and API key. Supports `/v1/chat/completions`, `/v1/embeddings`, and `/v1/images/generations` with full streaming, function calling, JSON mode, and vision inputs.

**WHAT MODELS ARE AVAILABLE?**  
The API includes frontier and open-source models such as Claude Opus-4.6, GPT-5.2, Gemini-3, Grok-4, Llama 4, Mistral, DeepSeek, and Qwen from leading providers. New models are added continuously as they launch.

**HOW IS PRICING STRUCTURED?**  
Pay per token with no per-seat fees. The free tier includes unlimited agent requests on Minimax-M2.5. Pro starts at $10/mo with $20 in credits, Pro Plus at $20/mo with $40 in credits, and Pro Max at $40/mo with $80 in credits. Enterprise volume discounts are available.

**DO YOU SUPPORT STREAMING?**  
Yes. Full SSE streaming support is available across all chat completion models. Response format matches the OpenAI spec exactly, so existing streaming implementations work without modification.

**WHAT SECURITY MEASURES DO YOU HAVE?**  
Infrastructure uses TLS 1.3 encryption in transit and AES-256 at rest. Enterprise plans add end-to-end encryption with zero-knowledge architecture, RBAC with SSO (Okta, Azure AD, Google Workspace), and comprehensive audit logging.

**WHAT IS THE API UPTIME GUARANTEE?**  
The API provides a 99.9% uptime SLA with sub-200ms median latency. Automatic fallback routing ensures availability even when individual providers experience outages, and smart load balancing optimizes for the lowest latency.

**CAN I USE THE API WITH EXISTING OPENAI SDKS?**  
Yes. The API is fully compatible with the official OpenAI Python and Node.js SDKs, as well as LangChain, LlamaIndex, and other popular frameworks. Migration takes under 5 minutes — just change the base URL.

### Start with BLACKBOX

Join 30M+ developers building with BLACKBOX AI.