Providers

Meridian supports multiple LLM providers through a single unified API. Add your provider keys once, and Meridian automatically routes requests to the correct provider based on the model name.

Adding a provider

  1. Navigate to the Providers page in the Meridian dashboard.
  2. Click Add Key on the provider card you want to configure.
  3. Paste your API key from the provider's console.
  4. Click Save. The key is encrypted with AES-256-GCM before storage.

You can add keys for all three providers simultaneously. Meridian detects which provider to use based on the model name in each request. The dashboard surfaces current recommended models instead of older or deprecated defaults.

How routing works

When a request arrives at any /api/v1/* endpoint, Meridian inspects the model field and matches it against known model prefixes:

Model prefixRoutes to
gpt-*, gpt-image-*, text-embedding-*, tts-*OpenAI
claude-*Anthropic
gemini-*Google Gemini

Responses are always translated to OpenAI-compatible format regardless of the upstream provider.

OpenAI

Get your API key from platform.openai.com/api-keys. Keys start with sk-.

ModelTypeDescription
gpt-4.1ChatCurrent flagship reasoning and generation model
gpt-4.1-miniChatBest balance of speed, quality, and cost
gpt-4.1-nanoChatLow-latency lightweight model
gpt-4oChatStrong multimodal model for general workloads
gpt-4o-miniChatAffordable multimodal option
gpt-image-1ImageCurrent OpenAI image generation model
text-embedding-3-smallEmbeddingSmall embedding model
text-embedding-3-largeEmbeddingLarge embedding model
tts-1AudioText-to-speech
tts-1-hdAudioHigh-definition text-to-speech

Anthropic

Get your API key from console.anthropic.com/settings/keys. Keys start with sk-ant-.

ModelTypeDescription
claude-sonnet-4-20250514ChatRecommended default for high-quality production workloads
claude-opus-4-20250514ChatHighest-capability Claude model
claude-3-7-sonnet-20250219ChatReliable cross-provider comparison baseline

Google Gemini

Get your API key from aistudio.google.com/app/apikey. Keys start with AIza.

ModelTypeDescription
gemini-2.5-proChatHighest-quality Gemini model
gemini-2.5-flashChatFast production default for most requests
gemini-2.0-flashChatLower-latency Gemini option
gemini-flash-latestChatRolling alias for the latest supported flash release

Removing a provider

Click Remove on the provider card in the dashboard. This deletes the encrypted key from the database. Any requests targeting that provider's models will return a 400 error until a new key is added.

Coming soon

Support for Mistral AI, Cohere, and Perplexity is planned for Meridian v2.0. See the Roadmap for details.