# Docs

- Ai studio
  - [Getting Started](/docs/ai-studio/getting-started): AI Studio: activate your account, explore projects and models, run your first inference in the Playground, then wire up programmatic access.
  - Concepts
    - [Organization](/docs/ai-studio/concepts/organization): Understand organizations, the top-level AI Studio platform for members, projects, billing, and cost
    - [Projects](/docs/ai-studio/concepts/projects): Understand AI Studio projects: members, API keys, model access, usage, and budgets
  - How to
    - [Choose a model](/docs/ai-studio/how-to/choose-a-model): Pick the right model for what you're building, then send your first request to it.
    - [Generate a training dataset](/docs/ai-studio/how-to/data-designer): Turn your documents into a high-quality training dataset in AI Studio. Covers supported files, size limits, and how NVIDIA NeMo Data Designer generates examples from your data.
    - [Fine-tune a model](/docs/ai-studio/how-to/fine-tuning): Train a LoRA adapter on your own JSONL dataset in AI Studio. Covers dataset format, job setup, hyperparameters, and what happens during training.
    - [Screen prompts with Guardrails](/docs/ai-studio/how-to/guardrails): Screen inference prompts for safety topics before they get to a model. This page tells you where the check runs, what applies to each project, the four outcomes, and the activity in AI Studio.
    - [Serve a fine-tuned model](/docs/ai-studio/how-to/serving): Load a finished LoRA adapter so Playground and API requests to it succeed. Covers adapter status, how long serve and unload take, and how to call the model.
  - Reference
    - [Reference](/docs/ai-studio/reference): Where to find AI Studio API details: endpoints, parameters, model identifiers, and error codes.
- Overview
  - [Overview](/docs/api-reference): The Nebul Inference API: an OpenAI-compatible (plus Anthropic Messages-compatible) proxy surface over the models your API key's project is authorized to use
  - Models: List and describe the various models available in the API.
    - [Models](/docs/api-reference/models)
    - [Model List](/docs/api-reference/models/model_list_v1_models_get)
    - [Model Info](/docs/api-reference/models/model_info_v1_model_info_get)
    - [Classify](/docs/api-reference/models/classify_v1_classify_post)
    - [Decisions](/docs/api-reference/models/decisions_v1_decisions_post)
    - [Get Model](/docs/api-reference/models/get_model_v1_models_get)
  - Chat: Given a list of messages comprising a conversation, the model will return a response.
    - [Chat](/docs/api-reference/chat)
    - [Chat Completion](/docs/api-reference/chat/chat_completion_v1_chat_completions_post)
    - [Create a Message (Anthropic-compatible)](/docs/api-reference/chat/create_message_v1_messages_post)
    - [Count Message Tokens](/docs/api-reference/chat/count_tokens_v1_messages_count_tokens_post)
  - Completions: The Chat Completions API endpoint will generate a model response from a list of messages comprising a conversation.
    - [Completions](/docs/api-reference/completions)
    - [Completion](/docs/api-reference/completions/completion_v1_completions_post)
  - Embeddings: Get a vector representation of a given input that can be easily consumed by machine learning models and algorithms.
    - [Embeddings](/docs/api-reference/embeddings)
    - [Embeddings](/docs/api-reference/embeddings/embeddings_v1_embeddings_post)
  - Images: Given a prompt and/or an input image, the model will generate or edit an image.
    - [Images](/docs/api-reference/images)
    - [Create Image](/docs/api-reference/images/create_image_v1_images_generations_post)
    - [Edit Image](/docs/api-reference/images/edit_image_v1_images_edits_post)
    - [Create Variation](/docs/api-reference/images/create_variation_v1_images_variations_post)
  - Audio: Turn audio into text or text into audio.
    - [Audio](/docs/api-reference/audio)
    - [Create Transcription](/docs/api-reference/audio/create_transcription_v1_audio_transcriptions_post)
    - [Create Speech](/docs/api-reference/audio/create_speech_v1_audio_speech_post)
    - [Audio Generations](/docs/api-reference/audio/audio_generations_v1_audio_generations_post)
    - [Transcribe Audio](/docs/api-reference/audio/transcribe_transcribe_post)
    - [Create Transcription (streaming)](/docs/api-reference/audio/v1_audio_transcriptions_stream_post)
    - [Create Transcription (stream input)](/docs/api-reference/audio/v1_audio_transcriptions_stream_input_post)
  - OCR: Extract text from documents using OCR. Supports PDF documents via URL or base64 content.
    - [OCR](/docs/api-reference/ocr)
    - [OCR Document](/docs/api-reference/ocr/ocr_v1_ocr_post)
  - Responses: Create and manage responses using OpenAI's Responses API. Supports structured outputs, function calling, and stateful conversations.
    - [Responses](/docs/api-reference/responses)
    - [Create Response](/docs/api-reference/responses/create_response_v1_responses_post)
  - Rerank: Rerank documents based on their relevance to a given query.
    - [Rerank](/docs/api-reference/rerank)
    - [Rerank Documents](/docs/api-reference/rerank/rerank_v1_rerank_post)
  - Files
    - [Files](/docs/api-reference/files)
    - [List Files](/docs/api-reference/files/list_files_files_v1_get)
    - [Upload File](/docs/api-reference/files/upload_file_files_v1_post)
    - [Get File](/docs/api-reference/files/get_file_files_v1_file_id_get)
    - [Delete File](/docs/api-reference/files/delete_file_files_v1_file_id_delete)
    - [Get File Content](/docs/api-reference/files/get_file_content_files_v1_file_id_content_get)
  - Meta
    - [Meta](/docs/api-reference/meta)
    - [List Changelog Entries](/docs/api-reference/meta/list_changelog_entries_changelog_entries_get)
- Inference api
  - Advanced topics
    - [Dedicated Endpoints API](/docs/inference-api/advanced-topics/dedicated-endpoints-api): Learn about dedicated inference endpoints with full model customizability and dedicated GPU resources
    - [Files API](/docs/inference-api/advanced-topics/files-api): Upload and manage files once, then reference them by ID across OCR, vision, and other Inference API requests
    - [Function Calling & Tools](/docs/inference-api/advanced-topics/function-calling-and-tools): Let models call external functions with structured arguments, and feed the results back into the conversation
    - [Prompt / KV Caching](/docs/inference-api/advanced-topics/prompt-kv-caching): Reuse repeated prompt context (KV cache) to reduce latency and cost, and isolate your cache from other customers with cache_salt
    - [Rate Limits & Scaling](/docs/inference-api/advanced-topics/rate-limits-and-scaling): How Nebul Inference API rate limits work, the headers returned, and how to handle 429 responses gracefully
    - [Reasoning Models](/docs/inference-api/advanced-topics/reasoning): Access model chain-of-thought reasoning traces alongside the final answer
    - [Request IDs](/docs/inference-api/advanced-topics/request-ids): Every Inference API response carries an X-Request-Id header for tracing and debugging. You can also supply your own.
    - [Speaker Diarization](/docs/inference-api/advanced-topics/speaker-diarization): Label different speakers in audio transcriptions for meetings, interviews, and conversations
    - [Structured Output & JSON](/docs/inference-api/advanced-topics/structured-output-and-json): Constrain model responses to valid JSON matching a schema you provide
    - [Voice Cloning](/docs/inference-api/advanced-topics/voice-cloning): Generate speech in a specific voice from a reference audio clip
  - Getting started
    - [Errors](/docs/inference-api/getting-started/errors): Every Inference API error code, where each comes from, exact response bodies, and retry guidance.
    - [Examples](/docs/inference-api/getting-started/examples): Runnable examples for streaming, embeddings, vision, function calling, and reasoning on the Inference API
    - [Introduction](/docs/inference-api/getting-started/introduction): What the Nebul Inference API is, what you can call, and where to start
    - [Quick Start](/docs/inference-api/getting-started/quick-start): Make your first request to the Nebul Inference API with the OpenAI SDKs or cURL
  - Models
    - [Chat Completions](/docs/inference-api/models/chat-completions): OpenAI-compatible chat completions endpoint for multi-turn conversations, system instructions, function calling, and structured outputs
    - [Embeddings](/docs/inference-api/models/embedding): Generate vector embeddings for semantic search, clustering, classification, and recommendations with the /v1/embeddings endpoint
    - [Image Models](/docs/inference-api/models/image-models): Generate, edit, and vary images from text prompts with the OpenAI-compatible Images API
    - [Messages API](/docs/inference-api/models/messages): Anthropic-compatible Messages API endpoint for conversational responses, tool use, streaming, and token counting
    - [Model Catalog](/docs/inference-api/models/model-catalog): Browse available Nebul Inference API models with their lifecycle stage, pricing, and metadata
    - [Model Lifecycle](/docs/inference-api/models/model-lifecycle): Lifecycle stages for models on the Nebul Inference API, how model version changes and deprecations are handled, and how to check them from the API.
    - [OCR](/docs/inference-api/models/ocr): Extract text from images and PDF documents into markdown with the OCR API
    - [Reranking](/docs/inference-api/models/reranking): Score and reorder documents by relevance to a query to improve search and RAG results
    - [Completions & Responses API](/docs/inference-api/models/responses): OpenAI-compatible text generation via the Responses endpoint, with reasoning and streaming, plus the legacy Completions endpoint
    - [Speech to Text](/docs/inference-api/models/speech-to-text): Transcribe audio to text with word- and segment-level timestamps, subtitle formats, streaming, and speaker diarization
    - [Text to Speech](/docs/inference-api/models/text-to-speech): Convert text to natural-sounding speech across 23 languages, with voice cloning and fine-grained delivery control
  - Resources
    - [Changelog](/docs/inference-api/resources/changelog): Release notes and updates for the Nebul Inference API
    - [Privacy & Security](/docs/inference-api/resources/privacy-and-security): How Nebul protects Inference API data: retention, EU residency, infrastructure security, and certifications
    - [Terms Of Service](/docs/inference-api/resources/terms-of-service): Usage terms and acceptable-use policies for the Nebul Inference API
- Integrations
  - [Claude Code](/docs/integrations/claude-code): Use Claude Code with the Nebul Inference API via Anthropic-compatible environment variables.
  - [Cline](/docs/integrations/cline): Use the Cline extension in VS Code with Nebul models through the OpenAI Compatible provider.
  - [Codex CLI](/docs/integrations/codex): Run OpenAI's Codex CLI on Nebul models through a custom model provider in config.toml.
  - [Continue](/docs/integrations/continue): Use the Continue extension in VS Code and JetBrains with Nebul models through the OpenAI-compatible provider.
  - [DeepSeek Harness](/docs/integrations/deepseek-harness): Run DeepSeek Harness (dsh) on Nebul models through a custom model API provider.
  - [Hermes Agent](/docs/integrations/hermes): Run the Hermes Agent from Nous Research on Nebul models through a custom endpoint.
  - [Kilo CLI](/docs/integrations/kilo-cli): Run the Kilo CLI terminal agent on Nebul models with an OpenAI-compatible custom provider.
  - [Kilo Code](/docs/integrations/kilo-code): Use Kilo Code in VS Code with the Nebul Inference API via the OpenAI-compatible provider.
  - [LiteLLM](/docs/integrations/litellm): Use Nebul models from LiteLLM, directly from the Python SDK or through a self-hosted LiteLLM proxy with budgets, virtual keys, and spend tracking.
  - [NemoClaw](/docs/integrations/nemoclaw): Run agents inside NVIDIA NemoClaw sandboxes with Nebul as the inference endpoint.
  - [OpenCode](/docs/integrations/open-code): Connect OpenCode to the Nebul Inference API using a custom OpenAI-compatible provider.
  - [OpenClaw](/docs/integrations/openclaw): Run OpenClaw on Nebul models through a custom provider with a base URL.
  - [Integrations Overview](/docs/integrations/overview): Connect any OpenAI- or Anthropic-compatible tool to the Nebul Inference API. Base URLs, model discovery, and telemetry opt-outs.
  - [Pi](/docs/integrations/pi): Run the Pi coding agent on Nebul models with an OpenAI-compatible endpoint in models.json.
  - [OpenAI SDKs and plain HTTP](/docs/integrations/sdks): Call the Nebul Inference API from Python, TypeScript, JavaScript, Go, Rust, Java, C++, and the shell using the OpenAI SDKs or a plain HTTP client.
  - [Zed](/docs/integrations/zed): Use Zed's agent panel with Nebul models through the OpenAI-compatible provider.
- Legal
  - [Acceptable Use Policy](/docs/legal/acceptable-use-policy): NeoCloud Acceptable Use Policy (AUP) v1.5.
  - [Customer Data Incident Response Process](/docs/legal/incident-response): Nebul Customer Data Incident Response Process v4.
  - [Master Agreement](/docs/legal/master-agreement): NeoCloud Master Agreement v1.6.
  - [Data Processing Addendum](/docs/legal/privacy-data-processing): NeoCloud Data Processing Addendum (DPA) v1.5.
  - [Nebul Services Privacy Policy](/docs/legal/services-privacy-policy): Nebul Services Privacy Policy v1.0.
  - [SLA — Infrastructure as a Service (IaaS)](/docs/legal/sla-iaas): NeoCloud Service Level Agreement for Infrastructure as a Service (IaaS) v1.4.
  - [SLA — Platform as a Service (PaaS)](/docs/legal/sla-paas): NeoCloud Service Level Agreement for Platform as a Service (PaaS) v1.5.
  - [SLA — Software as a Service (SaaS)](/docs/legal/sla-saas): NeoCloud Service Level Agreement for Software as a Service (SaaS) v1.1.
- Private neocloud
  - [Onboarding](/docs/private-neocloud/onboarding)
  - [VPN Access](/docs/private-neocloud/vpn-access)
  - Faq
    - [Developer Hub FAQ](/docs/private-neocloud/faq/developer-hub): Frequently asked questions about Developer Hub services on NeoCloud.
    - [Kubernetes FAQ](/docs/private-neocloud/faq/kubernetes): Frequently asked questions about Managed Kubernetes on NeoCloud.
    - [Managed Databases FAQ](/docs/private-neocloud/faq/managed-databases): Frequently asked questions about Managed Databases on NeoCloud.
    - [Network & Security FAQ](/docs/private-neocloud/faq/network-security): Frequently asked questions about networking and security services on NeoCloud.
    - [Open WebUI FAQ](/docs/private-neocloud/faq/open-webui): Frequently asked questions about Managed Open WebUI on NeoCloud.
    - [Storage FAQ](/docs/private-neocloud/faq/storage): Frequently asked questions about Storage services on NeoCloud.
  - Kubernetes
    - [Common Kubernetes Deployment Issues and How to Fix](/docs/private-neocloud/kubernetes/common-kubernetes-deployment-issues-and-how-to-fix)
    - [Creating a New Ingress Rule](/docs/private-neocloud/kubernetes/creating-a-new-ingress-rule)
    - [How do I properly size CPU and memory for my workloads?](/docs/private-neocloud/kubernetes/how-do-i-properly-size-cpu-and-memory-for-my-workloads)
    - [How to fix if POD that are OOM Killed](/docs/private-neocloud/kubernetes/how-to-fix-if-pod-that-are-oom-killed)
    - [How to get access to K8s clusters](/docs/private-neocloud/kubernetes/how-to-get-access-to-k8s-clusters)
    - [Nvidia Gpus](/docs/private-neocloud/kubernetes/nvidia-gpus)
    - [Using Cert-Manager](/docs/private-neocloud/kubernetes/using-cert-manager)
    - [Using Loadbalancers](/docs/private-neocloud/kubernetes/using-loadbalancers)
  - Service descriptions
    - [Introduction](/docs/private-neocloud/service-descriptions/service-neocloud)
    - Infrastructure
      - [Compute](/docs/private-neocloud/service-descriptions/infrastructure/compute)
      - [Landing Zone & Network](/docs/private-neocloud/service-descriptions/infrastructure/landingzone-and-network)
      - [Storage](/docs/private-neocloud/service-descriptions/infrastructure/storage)
    - Platform
      - [Developer Hub](/docs/private-neocloud/service-descriptions/platform/developer-hub): Developer platform services on NeoCloud.
      - [Kubernetes](/docs/private-neocloud/service-descriptions/platform/kubernetes): Managed Kubernetes — container orchestration on NeoCloud.
      - [Managed Databases](/docs/private-neocloud/service-descriptions/platform/managed-databases): Managed database services on NeoCloud — PostgreSQL, Qdrant, Valkey, and MongoDB (FerretDB).
      - [Network & Security](/docs/private-neocloud/service-descriptions/platform/network-security): Managed networking and security services on NeoCloud.
      - [Open WebUI](/docs/private-neocloud/service-descriptions/platform/open-webui): Managed Open WebUI — AI chat interface on NeoCloud.
      - [OpenBao](/docs/private-neocloud/service-descriptions/platform/openbao): Managed OpenBao — secrets management on NeoCloud.
  - Service Terms
    - [Service Terms](/docs/private-neocloud/service-terms): NeoCloud service-specific terms and conditions.
    - [Managed Kubernetes — Service Terms](/docs/private-neocloud/service-terms/kubernetes): Service-specific terms for Managed Kubernetes on NeoCloud.
    - [Managed Databases — Service Terms](/docs/private-neocloud/service-terms/managed-databases): Service-specific terms for Managed Databases on NeoCloud.
    - [Network & Security — Service Terms](/docs/private-neocloud/service-terms/network-security): Service-specific terms for Network & Security services on NeoCloud.
    - [Open WebUI — Service Terms](/docs/private-neocloud/service-terms/open-webui): Service-specific terms for Managed Open WebUI on NeoCloud.
    - [OpenBao — Service Terms](/docs/private-neocloud/service-terms/openbao): Service-specific terms for Managed OpenBao on NeoCloud.
    - [Storage — Service Terms](/docs/private-neocloud/service-terms/storage): Service-specific terms for Storage services on NeoCloud.
- Support
  - Getting started
    - [Before You Submit a Ticket](/docs/support/getting-started/before-you-submit): Prepare for the fastest resolution by gathering the right information before submitting a support ticket.
    - [How Nebul Support Works](/docs/support/getting-started/how-support-works): Learn how Nebul Support operates and what to expect when you raise a ticket.
  - Guides
    - [Getting the Fastest Resolution](/docs/support/guides/fastest-resolution): Learn what information to provide for the fastest support resolution, with specific guidance for Inference API and Private NeoCloud issues.
    - [Opening a Support Ticket](/docs/support/guides/opening-a-ticket): Step-by-step guide to submitting a support ticket and what information to include for the fastest resolution.
  - Plans
    - [Support Plans](/docs/support/plans/overview): Compare Nebul support plans - Basic, Tech, and Relentless - and find the right tier for your operational needs.