Multi-Provider Routing
ModelHive provides a single, unified API that gives you access to models across all major AI providers. Switch models by changing one string — no SDK changes, no new credentials, no new billing relationships.
One API, Every Model
# OpenAI GPT-4o
curl https://api.modelhive.ai/v1/chat/completions \
-H "Authorization: Bearer sk-your-key" \
-d '{"model": "gpt-4o", "messages": [...]}'
# Anthropic Claude (same API, same key)
curl https://api.modelhive.ai/v1/chat/completions \
-H "Authorization: Bearer sk-your-key" \
-d '{"model": "vertex-claude-sonnet-4-20250514", "messages": [...]}'
# Google Gemini (same API, same key)
curl https://api.modelhive.ai/v1/chat/completions \
-H "Authorization: Bearer sk-your-key" \
-d '{"model": "vertex-gemini-2.0-flash", "messages": [...]}'
Supported Providers
| Provider | Models | Data Residency |
|---|---|---|
| OpenAI | GPT-4o, GPT-4o-mini, GPT-5, o3, o4-mini | US |
| Anthropic | Claude Sonnet 4, Claude Haiku 3.5 | EU |
| Google Gemini | Gemini 2.0 Flash, Gemini 2.5 Pro | EU |
| Azure OpenAI | GPT-4.1, GPT-5, Codex, DALL-E, Whisper, Sora | EU |
| Mistral | Mistral Large, Codestral | EU |
| DeepSeek | DeepSeek-R1, DeepSeek-V3 | EU |
| Meta | Llama 4 Maverick, Llama 4 Scout | EU |
| xAI | Grok | EU |
| Cohere | Command R+ | EU |
Plus image generation (DALL-E, Imagen, FLUX), video (Sora, Veo), audio (Whisper, TTS), and embeddings.
EU Data Residency
ModelHive is designed with GDPR compliance as a first-class feature:
- EU models are hosted in European data centers (Belgium, Sweden)
- Azure models run in
swedencentral(Sweden) - Only OpenAI direct routes to US data centers
- Analytics track EU vs. non-EU request volumes automatically
Per-Tenant Model Access
Admins can control exactly which models each tenant has access to from the ModelHive Dashboard → Tenant Settings → Models. Select the models you want to enable for each tenant — users will only see and be able to use the models you've approved.
OpenAI-Compatible
ModelHive uses the OpenAI API format — so any SDK, tool, or library that works with OpenAI works with ModelHive. Just change the base URL:
from openai import OpenAI
client = OpenAI(
api_key="sk-your-modelhive-key",
base_url="https://api.modelhive.ai/v1"
)
All requests go through the same HiveCache → Guardrails → Compression pipeline regardless of the provider — so you get caching, security, and cost optimization on every model.