Skip to main content

Models & Providers

ModelHive exposes every model through a single OpenAI-compatible API. You call a model by its model id — the same value you pass as "model" in a request and the same value returned by GET /v1/models.

The Data Residency column tells you where inference happens:

  • EU — the request is served by a European deployment (Azure Data Zone or regional EU deployment, AWS Bedrock eu-* inference profiles, Vertex AI europe-west*). Prompts and completions stay inside the EU.
  • Global — the provider may route the request outside the EU.

For a model with several deployments, ModelHive fails over between them; the model counts as EU only when the deployment actually serving the request is European.

Model Catalogue​

ModelModel IDProviderInfrastructureData Residency
Qwen3 Embedding 0.6Bqwen3-embeddingsAlibabaSelf-hostedGlobal
Nova 2 Litenova-2-liteAmazonAWS BedrockEU
Nova Litenova-liteAmazonAWS BedrockEU
Nova Micronova-microAmazonAWS BedrockEU
Nova Pronova-proAmazonAWS BedrockEU
Claude Fable 5claude-fable-5AnthropicGoogle Vertex AIGlobal
Claude Haiku 4.5vertex-claude-haiku-4-5AnthropicAWS Bedrock, Google Vertex AIEU
Claude Opus 4.5vertex-claude-opus-4-5AnthropicAWS Bedrock, Google Vertex AIEU
Claude Opus 4.6claude-opus-4-6AnthropicAWS Bedrock, Azure AI Foundry, Google Vertex AIEU / Global
Claude Opus 4.7claude-opus-4-7AnthropicAzure AI FoundryGlobal
Claude Opus 4.8claude-opus-4-8AnthropicAzure AI FoundryGlobal
Claude Opus 5claude-opus-5AnthropicAzure AI Foundry, Google Vertex AIGlobal
Claude Sonnet 4.5vertex-claude-sonnet-4-5AnthropicAWS Bedrock, Google Vertex AIEU
Claude Sonnet 4.6claude-sonnet-4-6AnthropicAWS Bedrock, Azure AI Foundry, Google Vertex AIEU / Global
Claude Sonnet 5claude-sonnet-5AnthropicGoogle Vertex AIGlobal
FLUX 1.1 ProFLUX-1.1-proBlack Forest LabsAzure AI FoundryEU
FLUX.1 Kontext ProFLUX.1-Kontext-proBlack Forest LabsAzure AI FoundryEU
FLUX.2 FlexFLUX.2-flexBlack Forest LabsAzure AI FoundryGlobal
FLUX.2 ProFLUX.2-proBlack Forest LabsAzure AI FoundryEU
Cohere Command A Plus (05-2026)Cohere-command-a-plus-05-2026CohereAzure AI FoundryGlobal
Cohere Embed 4.0embed-v-4-0CohereAzure AI FoundryGlobal
Cohere Embed V4cohere-embed-v4CohereAWS BedrockEU
Cohere Parse V5Cohere-parse-v5CohereAzure AI FoundryGlobal
Cohere Rerank V4.0 FastCohere-rerank-v4.0-fastCohereAzure AI FoundryGlobal
Cohere Rerank V4.0 ProCohere-rerank-v4.0-proCohereAzure AI FoundryGlobal
DeepSeek R1DeepSeek-R1DeepSeekAzure AI FoundryGlobal
DeepSeek V3.2DeepSeek-V3.2DeepSeekAzure AI FoundryGlobal
DeepSeek V3.2 SpecialeDeepSeek-V3.2-SpecialeDeepSeekAzure AI FoundryGlobal
DeepSeek V4 FlashDeepSeek-V4-FlashDeepSeekAzure AI FoundryEU
DeepSeek V4 ProDeepSeek-V4-ProDeepSeekAzure AI FoundryGlobal
Gemini 2.5 Flashgemini-2.5-flashGoogleGoogle Vertex AIEU
Gemini 2.5 Flash Litegemini-2.5-flash-liteGoogleGoogle Vertex AIEU
Gemini 2.5 Progemini-2.5-proGoogleGoogle Vertex AIEU
Gemini 3 Flash Previewgemini-3-flash-previewGoogleGoogle Vertex AIGlobal
Gemini 3.1 Flash Litegemini-3.1-flash-liteGoogleGoogle Vertex AIGlobal
Gemini 3.1 Pro Previewgemini-3.1-pro-previewGoogleGoogle Vertex AIGlobal
Gemini 3.5 Flashgemini-3.5-flashGoogleGoogle Vertex AIEU
Gemini 3.5 Flash Litegemini-3.5-flash-liteGoogleGoogle Vertex AIGlobal
Gemini 3.6 Flashgemini-3.6-flashGoogleGoogle Vertex AIGlobal
Gemini 3.7 Flashgemini-3.7-flashGoogleGoogle Vertex AIGlobal
Gemini 3.8 Flashgemini-3.8-flashGoogleGoogle Vertex AIGlobal
Llama 4 Maverick 17B 128E Instruct FP8Llama-4-Maverick-17B-128E-Instruct-FP8MetaAzure AI FoundryGlobal
Llama 4 Scout 17B 16E InstructLlama-4-Scout-17B-16E-InstructMetaAzure AI FoundryGlobal
MAI Image 2.6MAI-Image-2.6MicrosoftAzure AI FoundryGlobal
MAI Image 2.6 FlashMAI-Image-2.6-FlashMicrosoftAzure AI FoundryGlobal
MAI Thinking 1MAI-Thinking-1MicrosoftAzure AI FoundryGlobal
Model Routermodel-routerMicrosoftAzure OpenAIEU
MiniMax M2.5minimaxMiniMaxOpenRouterGlobal
Mistral Document AI 2512mistral-document-ai-2512Mistral AIAzure AI FoundryEU
Mistral Large 3Mistral-Large-3Mistral AIAzure AI FoundryEU
Mistral Medium 2505mistral-medium-2505Mistral AIAzure AI FoundryGlobal
Mistral Medium 3.5mistral-medium-3-5Mistral AIAzure AI FoundryEU
Mistral OCR 4.0mistral-ocr-4-0Mistral AIAzure AI FoundryEU
Mistral Small 2503mistral-small-2503Mistral AIAzure AI FoundryGlobal
Pixtral Largepixtral-largeMistral AIAWS BedrockEU
Kimi K2.5Kimi-K2.5Moonshot AIAzure AI FoundryGlobal
Kimi K2.6Kimi-K2.6Moonshot AIAzure AI FoundryGlobal
Nemotron 3 UltranemotronNVIDIAOpenRouterGlobal
GPT 4ogpt-4oOpenAIAzure OpenAIEU
GPT 4o Minigpt-4o-miniOpenAIAzure OpenAIEU
GPT 5 Minigpt-5-miniOpenAIAzure OpenAIEU
GPT 5 Nanogpt-5-nanoOpenAIAzure OpenAIEU
GPT 5.1gpt-5.1OpenAIAzure OpenAIEU
GPT 5.1 Chatgpt-5.1-chatOpenAIAzure OpenAIGlobal
GPT 5.1 Codex Maxgpt-5.1-codex-maxOpenAIAzure OpenAIGlobal
GPT 5.2 Chatgpt-5.2-chatOpenAIAzure OpenAIGlobal
GPT 5.2 Codexgpt-5.2-codexOpenAIAzure OpenAIGlobal
GPT 5.3 Chatgpt-5.3-chatOpenAIAzure OpenAIGlobal
GPT 5.3 Codexgpt-5.3-codexOpenAIAzure OpenAIGlobal
GPT 5.4gpt-5.4OpenAIAzure OpenAIGlobal
GPT 5.4 Minigpt-5.4-miniOpenAIAzure OpenAIEU
GPT 5.4 Nanogpt-5.4-nanoOpenAIAzure OpenAIGlobal
GPT 5.4 Progpt-5.4-proOpenAIAzure OpenAIGlobal
GPT 5.5gpt-5.5OpenAIAzure OpenAIEU
GPT 5.6 Lunagpt-5.6-lunaOpenAIAzure OpenAIEU
GPT 5.6 Solgpt-5.6-solOpenAIAzure OpenAIEU
GPT 5.6 Terragpt-5.6-terraOpenAIAzure OpenAIEU
GPT 6 Astragpt-6-astraOpenAIAzure OpenAIGlobal
GPT Audio 1.5gpt-audio-1.5OpenAIAzure OpenAIGlobal
GPT Image 2gpt-image-2OpenAIAzure OpenAIGlobal
GPT Live Transcribegpt-live-transcribeOpenAIAzure OpenAIGlobal
GPT Realtime 1.5gpt-realtime-1.5OpenAIAzure OpenAIGlobal
GPT Realtime 2.1gpt-realtime-2.1OpenAIAzure OpenAIEU
GPT Transcribegpt-transcribeOpenAIAzure OpenAIGlobal
GPT-OSS 120Bgpt-ossOpenAIAzure OpenAI, Cerebras, Groq, OpenRouterGlobal
GPT-OSS 20Bgpt-oss:20bOpenAIGroqGlobal
Text Embedding 3 Largetext-embedding-3-largeOpenAIAzure AI FoundryEU
Text Embedding 3 Smalltext-embedding-3-smallOpenAIAzure AI FoundryEU
WhisperwhisperOpenAIAzure OpenAIEU
Grok 4.1 Fast Non Reasoninggrok-4-1-fast-non-reasoningxAIAzure AI FoundryGlobal
Grok 4.1 Fast Reasoninggrok-4-1-fast-reasoningxAIAzure AI FoundryGlobal
Grok 4.20 Non Reasoninggrok-4-20-non-reasoningxAIAzure AI FoundryGlobal
Grok 4.20 Reasoninggrok-4-20-reasoningxAIAzure AI FoundryGlobal
Grok 4.6grok-4.6xAIAzure AI FoundryGlobal

Listing System Models​

curl -X GET https://api.modelhive.ai/api/models \
  -H "Authorization: Bearer {jwt_token}"

Response:

[
  {
    "id": "gpt-4o",
    "name": "GPT 4o",
    "provider": "openai",
    "is_eu": true,
    "description": null
  },
  {
    "id": "gemini-2.5-pro",
    "name": "Gemini 2.5 Pro",
    "provider": "google",
    "is_eu": true,
    "description": null
  }
]

Tenant Models​

Admins can control which models each tenant has access to from the ModelHive Dashboard → Tenant Settings → Models. Select the models you want to enable — users will only see and be able to use the approved models.

Model Naming Convention​

Most models are exposed under a short, vendor-neutral id. When the same model is available from more than one cloud, the short id is the multi-provider entry and ModelHive fails over between deployments; the prefixed ids let you pin a specific one.

FormatExampleDescription
model-namegpt-4o, gemini-2.5-proStandard id, ModelHive picks the deployment
azure-modelazure-claude-sonnet-4-6Pinned to Azure AI Foundry
bedrock-modelbedrock-claude-sonnet-4-6Pinned to AWS Bedrock
vertex-modelvertex-claude-sonnet-4-6Pinned to Google Vertex AI

EU Data Residency​

ModelHive tracks which models process data within the EU for GDPR compliance:

import requests

response = requests.get(
    "https://api.modelhive.ai/api/models",
    headers={"Authorization": f"Bearer {jwt_token}"}
)

eu_models = [m for m in response.json() if m.get("is_eu")]
print(f"EU-compliant models: {[m['id'] for m in eu_models]}")

Per-Key Model Restrictions​

When creating an API key from the Dashboard, you can restrict it to specific models. This is useful for isolating keys for specific use cases (e.g., a key that can only use EU-compliant models).