AI Gateway

Claude

Opus, Sonnet, Haiku — best-in-class reasoning, analysis, and code generation. Premium tier default for complex tasks.

Gemini

Pro, Flash — multimodal understanding and long-context document processing. Efficient at high token volumes.

GPT-4o

Vision, function calling, and broad general capabilities. Widely compatible with existing integrations.

Mistral

Codestral, Large — EU data sovereignty and code generation. Routes EU-origin requests to Mistral by default for GDPR compliance.

Grok

Real-time knowledge and structured reasoning. Suitable for time-sensitive queries and current-events context.

Open Models

Llama, Qwen, DeepSeek — community-maintained open models for economy-tier tasks and research workloads.

Smart Routing

Automatic model selection based on task complexity, prompt length, and cost targets. Routes simple tasks to economy models, complex tasks to premium.

Cost Optimization

Tier-based pricing with economy, standard, and premium routing. Daily and weekly spend caps enforced per account.

Unified API

One endpoint for all providers. Consistent request and response format regardless of the underlying model. No per-provider SDK required.

POST /api/v1/claude/complete

EU Compliance

Requests flagged as EU-origin are automatically routed to Mistral infrastructure. GDPR-compliant by default, no configuration needed.

Claude

POST /api/v1/claude/complete

Generate, summarize, analyze, translate, rewrite. Supports streaming and multi-turn context.

Gemini

POST /api/v1/gemini/generate

Multimodal generation with image and document input. Long-context support up to 1M tokens.

Mistral

POST /api/v1/mistral/generate

EU-sovereign generation and code completion. Codestral variant available for developer tasks.

Open Models

POST /api/v1/openrouter/generate

Access to 100+ open and community models via OpenRouter. Economy tier default.