# concentrate

## Docs

- [Anomalous spend](https://concentrate.ai/docs/alerts/anomalous-spend.md): Get notified when spending suddenly deviates from your normal pattern
- [Balance depletion](https://concentrate.ai/docs/alerts/balance-depletion.md): Get a heads-up when your balance is trending toward zero
- [Dashboard snapshot](https://concentrate.ai/docs/alerts/dashboard-snapshot.md): Get a recurring summary of your account at a glance
- [Dormant key](https://concentrate.ai/docs/alerts/dormant-key.md): Get notified when a long-inactive API key starts being used again
- [Error rate spike](https://concentrate.ai/docs/alerts/error-rate-spike.md): Get notified when your error rate climbs above a threshold
- [Key exhaustion](https://concentrate.ai/docs/alerts/key-exhaustion.md): Get a heads-up when an API key is forecast to hit its usage cap soon
- [Key limit](https://concentrate.ai/docs/alerts/key-limit.md): Get notified immediately when an API key crosses a percentage of its limit
- [Low balance](https://concentrate.ai/docs/alerts/low-balance.md): Get notified immediately when your balance drops below a dollar threshold
- [Model usage report](https://concentrate.ai/docs/alerts/model-usage-report.md): Get a recurring breakdown of which models and providers are costing you the most
- [Alerts overview](https://concentrate.ai/docs/alerts/overview.md): What alerts are, when to use each one, and how they reach you
- [Auto routing and model fallback](https://concentrate.ai/docs/api-reference/endpoint/auto-routing.md): Configure auto model selection, provider fallback, cost or performance routing, graceful degradation, and ZDR enforcement on the Responses API.
- [Bring Your Own Key (BYOK)](https://concentrate.ai/docs/api-reference/endpoint/byok.md): Store your own provider API keys and have Concentrate route requests through them — no rate limits, no credit spend, and if your keys don't work your requests still go through.
- [Chat Completions](https://concentrate.ai/docs/api-reference/endpoint/chat-completions.md): OpenAI Chat Completions API compatibility endpoint for Cursor, Opencode, and other clients
- [Create Message](https://concentrate.ai/docs/api-reference/endpoint/create-message.md): Anthropic Messages API compatibility endpoint for Claude Code and other clients
- [Create Response](https://concentrate.ai/docs/api-reference/endpoint/create-response.md): Start a stateful conversation with your model of choice
- [Error Handling](https://concentrate.ai/docs/api-reference/endpoint/errors.md): Complete guide to API error responses and troubleshooting
- [Get Model](https://concentrate.ai/docs/api-reference/endpoint/get-model.md): Get details for a specific model
- [Get Provider Info](https://concentrate.ai/docs/api-reference/endpoint/get-provider-info.md): Get provider-specific information for a model
- [Guardrails and PII redaction for the Responses API](https://concentrate.ai/docs/api-reference/endpoint/guardrails-redaction.md): Redact PII from prompts and responses by configuring key-level guardrails or calling the redact-v1 model directly on /v1/responses.
- [Health Check](https://concentrate.ai/docs/api-reference/endpoint/health.md)
- [List Models](https://concentrate.ai/docs/api-reference/endpoint/list-models.md): Retrieve the complete catalog of AI models
- [List Models by Author](https://concentrate.ai/docs/api-reference/endpoint/list-models-by-author.md): Retrieve all models from a specific author
- [List Models by Provider](https://concentrate.ai/docs/api-reference/endpoint/list-models-by-provider.md): Retrieve all models available from a specific provider
- [List Providers](https://concentrate.ai/docs/api-reference/endpoint/list-providers.md): List all available providers information
- [Multi-Modal Inputs](https://concentrate.ai/docs/api-reference/endpoint/multi-modal.md): Send image content alongside text to vision-capable models on the Concentrate AI API, including supported formats, size and resolution limits, and examples.
- [Prompt Caching](https://concentrate.ai/docs/api-reference/endpoint/prompt-caching.md): Reduce latency and input cost by reusing stable prompt prefixes
- [Request Parameters Reference](https://concentrate.ai/docs/api-reference/endpoint/request-parameters.md): Reference for every request parameter on the Responses API, including model, input, routing, tools, sampling, streaming, and response format options.
- [Streaming Responses](https://concentrate.ai/docs/api-reference/endpoint/streaming.md): Real-time streaming using Server-Sent Events
- [Structured Output](https://concentrate.ai/docs/api-reference/endpoint/structured-output.md): Force models to return JSON matching a specific schema
- [Supported AI models catalog and provider availability](https://concentrate.ai/docs/api-reference/endpoint/supported-models.md): Browse 150+ AI models on Concentrate by author with model slugs, context windows, max output tokens, and per-provider availability for routing.
- [Tool Calling](https://concentrate.ai/docs/api-reference/endpoint/tool-calling.md): Enable models to call functions and use external tools
- [Web Search](https://concentrate.ai/docs/api-reference/endpoint/web-search.md): Give models access to real-time web search results
- [Zero Data Retention & Logging](https://concentrate.ai/docs/api-reference/endpoint/zero-data-retention.md): Configure Zero Data Retention and request logging on Concentrate API keys, and see which models and providers offer ZDR-certified inference endpoints.
- [API Introduction](https://concentrate.ai/docs/api-reference/introduction.md): Unified API for accessing 150+ AI models through a single interface
- [Quickstart](https://concentrate.ai/docs/getting-started/quickstart.md): Make your first API call in under 5 minutes
- [Claude Code](https://concentrate.ai/docs/integrations/claude-code.md): Setup Claude Code in 30 seconds and code with 50+ models including MiniMax, Kimi, GLM, DeepSeek, OpenAI, Gemini, and Claude. No expensive $200 subscription required. Get a Concentrate API key and pay as you go.
- [Claude Desktop](https://concentrate.ai/docs/integrations/claude-desktop.md): Route Claude Desktop's Cowork through Concentrate using the third-party platform gateway.
- [Codex (Beta)](https://concentrate.ai/docs/integrations/codex.md): Use models on Concentrate with the Codex CLI
- [Cursor (Beta)](https://concentrate.ai/docs/integrations/cursor.md): Use models on Concentrate in Cursor via OpenAI-compatible settings
- [Integrations overview](https://concentrate.ai/docs/integrations/overview.md): Connect Concentrate AI to Claude Code, Claude Desktop, Codex, Cursor, and other OpenAI-compatible tools with a base URL and API key.
- [Helicone](https://concentrate.ai/docs/migrations/helicone.md): Migrate from Helicone to Concentrate AI: what changes, what carries over, and what does not.
- [LiteLLM](https://concentrate.ai/docs/migrations/litellm.md): Migrate from LiteLLM to Concentrate AI: what changes, what carries over, and what does not.
- [Merge](https://concentrate.ai/docs/migrations/merge.md): Migrate from Merge Gateway to Concentrate AI: what changes, what carries over, and what does not.
- [OpenRouter](https://concentrate.ai/docs/migrations/openrouter.md): Migrate from OpenRouter to Concentrate AI with minimal code changes.
- [Migrations overview](https://concentrate.ai/docs/migrations/overview.md): Move an existing LLM integration to Concentrate AI with minimal code changes
- [Portkey](https://concentrate.ai/docs/migrations/portkey.md): Migrate from Portkey to Concentrate AI: what changes, what carries over, and what does not.
- [TensorZero](https://concentrate.ai/docs/migrations/tensorzero.md): Migrate from TensorZero to Concentrate AI: what changes, what carries over, and what does not.
- [Migrate from Vercel AI Gateway](https://concentrate.ai/docs/migrations/vercel.md): Move from Vercel AI Gateway to Concentrate AI: code changes for the AI SDK and OpenAI client, model IDs, auth, and feature differences.

## OpenAPI Specs

- [openapi](https://concentrate.ai/docs/api-reference/openapi.json)