# Subconscious Docs

- [Subconscious](https://docs.subconscious.dev/overview.md)
- [Quickstart](https://docs.subconscious.dev/quickstart.md): Make your first API call
- [Streaming](https://docs.subconscious.dev/features/streaming.md): Stream responses token by token in real time
- [Thinking Mode](https://docs.subconscious.dev/features/thinking.md): Enable step-by-step reasoning for complex tasks
- [Subconscious Cache](https://docs.subconscious.dev/features/subconscious-cache.md): (Advanced) Disable auto compaction and trigger the subconscious cache manually.
- [Model discovery](https://docs.subconscious.dev/features/model-discovery.md): Discover available models, their capabilities, and their limits
- [Management API keys](https://docs.subconscious.dev/features/management-api-keys.md): Provision and manage inference keys programmatically
- [Session affinity](https://docs.subconscious.dev/features/session-affinity.md): Keep a conversation's follow-up requests on the worker that already holds its KV cache, so cached context is reused instead of recomputed.
- [Request correlation](https://docs.subconscious.dev/features/request-correlation.md): Group requests into conversations and label your client in usage views, so multi-turn traffic is readable turn by turn.
- [Subagent correlation](https://docs.subconscious.dev/features/subagent-correlation.md): Nest a spawned agent's conversation under the conversation that spawned it, so parallel agent work stays readable.
- [API](https://docs.subconscious.dev/ways-to-use/cloud-api.md): The standard Subconscious inference endpoint
- [Dedicated Endpoints](https://docs.subconscious.dev/ways-to-use/dedicated.md): Reserved compute with guaranteed capacity
- [On-prem and customer cloud](https://docs.subconscious.dev/ways-to-use/on-prem.md): Deploy Subconscious inference in your own infrastructure
- [Privacy and security](https://docs.subconscious.dev/ways-to-use/privacy.md): How we handle and protect your data
- [Learn More](https://docs.subconscious.dev/resources/learn-more.md): Explore more about our work and our team
- [List models](https://docs.subconscious.dev/api-reference/models.md): Returns the model catalog for the caller. Each entry describes the model's lifecycle, limits, modalities, capabilities, supported request wires, deployment facts, and the rate limits enforced for this caller on that model.
- [List models with pricing](https://docs.subconscious.dev/api-reference/platform-models.md): List the caller's models with per-token prices and partner formats
- [Create Chat Completion](https://docs.subconscious.dev/api-reference/chat-completions.md): Generates a model response for the given conversation. This endpoint is fully compatible with the OpenAI Chat Completions API.
- [Create Message](https://docs.subconscious.dev/api-reference/messages.md): Generates a model response for the given conversation. This endpoint is fully compatible with the Anthropic Messages API, so you can use the Anthropic SDK by pointing its base URL at Subconscious.
- [Create Response](https://docs.subconscious.dev/api-reference/responses.md): Generates a model response from an input sequence. This endpoint is compatible with the OpenAI Responses API, which Codex and the Vercel AI SDK use. Session affinity works the same as on chat completions: a body session id, a session header, or prompt_cache_key groups requests into one conversation.
- [Create inference key](https://docs.subconscious.dev/api-reference/create-inference-key.md): Mint an inference key with a management key
- [List inference keys](https://docs.subconscious.dev/api-reference/list-inference-keys.md): List inference keys for the management key's organization
- [Delete inference key](https://docs.subconscious.dev/api-reference/delete-inference-key.md): Revoke an inference key by its prefix
- [Get usage](https://docs.subconscious.dev/api-reference/usage.md): Read today's organization usage with an inference key
- [Overview](https://docs.subconscious.dev/on-prem/overview.md): What is the Subconscious Inference System, what are the benefits, and who is it for
- [How it works](https://docs.subconscious.dev/on-prem/how-it-works.md): The API Gateway, OrangeLine, and Ryvn behind the Subconscious Inference System
- [FAQ](https://docs.subconscious.dev/on-prem/faq.md): Common questions about the Subconscious Inference System
- [Overview](https://docs.subconscious.dev/on-prem/integration-journey/overview.md): How evaluation, onboarding, and long-term support work
- [Evaluation](https://docs.subconscious.dev/on-prem/integration-journey/evaluation.md): Discovery, ROI, security review, and optional trials
- [Onboarding](https://docs.subconscious.dev/on-prem/integration-journey/onboarding.md): From kickoff to pilot users on the Subconscious Inference System
- [Customer success](https://docs.subconscious.dev/on-prem/integration-journey/customer-success.md): Support, upgrades, reviews, and expansion after onboarding
- [Compliance](https://docs.subconscious.dev/on-prem/trust-center/compliance.md): Security review packet for the customer-hosted Subconscious Inference System
- [Methods](https://docs.subconscious.dev/on-prem/deployments/methods.md): Full inference system via Ryvn, or OrangeLine only as a Docker container
- [Configurations](https://docs.subconscious.dev/on-prem/deployments/configurations.md): Where OrangeLine GPUs run relative to a BYOC API Gateway
- [Overview](https://docs.subconscious.dev/on-prem/api-gateway/overview.md): The customer-facing control point for coding-agent traffic
- [Setup](https://docs.subconscious.dev/on-prem/api-gateway/setup.md): Admin setup for dashboard access, routes, credentials, and pilot readiness
- [OrangeLine](https://docs.subconscious.dev/on-prem/inference-runtime/overview.md): GPU inference runtime you can run anywhere a Docker container will start
- [Overview](https://docs.subconscious.dev/on-prem/ryvn/overview.md): Ryvn deploys and updates the API Gateway in your AWS, GCP, or Azure account

## OpenAPI Specs

- [openapi](/api-reference/openapi.json)
- [platform](/api-reference/platform.json)


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.