# Traceport

## Docs

- [Playground](https://docs.traceport.ai/ai-tools/playground.md): Test and compare models side-by-side with an interactive prompt editor.
- [Cancel Batch](https://docs.traceport.ai/api-reference/endpoint/cancel-batch.md): Cancels an in-progress batch.
- [Create Batch](https://docs.traceport.ai/api-reference/endpoint/create-batch.md): Creates a batch of API requests for asynchronous processing.
- [Create Chat Completion](https://docs.traceport.ai/api-reference/endpoint/create-chat-completion.md): Creates a model response for the given chat conversation. Supports OpenAI, Anthropic, Google, and Bedrock providers via provider resolution based on the model name. Set `stream: true` for Server-Sent Events streaming.
- [Create Embeddings](https://docs.traceport.ai/api-reference/endpoint/create-embeddings.md): Creates an embedding vector representing the input text.
- [Delete File](https://docs.traceport.ai/api-reference/endpoint/delete-file.md): Deletes a file.
- [Execute Workflow](https://docs.traceport.ai/api-reference/endpoint/execute-workflow.md): Executes a server-side workflow by its config version ID.
- [List Batches](https://docs.traceport.ai/api-reference/endpoint/list-batches.md): Returns a list of batches.
- [List Files](https://docs.traceport.ai/api-reference/endpoint/list-files.md): Returns a list of files that belong to the user's organization.
- [Retrieve Batch](https://docs.traceport.ai/api-reference/endpoint/retrieve-batch.md): Retrieves a batch by its ID.
- [Retrieve File](https://docs.traceport.ai/api-reference/endpoint/retrieve-file.md): Returns information about a specific file.
- [Retrieve File Content](https://docs.traceport.ai/api-reference/endpoint/retrieve-file-content.md): Returns the raw content of the specified file.
- [Run Prompt](https://docs.traceport.ai/api-reference/endpoint/run-prompt.md): Runs the published version of a prompt template identified by its slug. Template variables in `{{variable}}` format are resolved from the `variables` map in the request body.
- [Upload File](https://docs.traceport.ai/api-reference/endpoint/upload-file.md): Uploads a file for use with batch processing or other features.
- [API Reference](https://docs.traceport.ai/api-reference/introduction.md): Traceport Gateway API — OpenAI-compatible unified gateway for multiple LLM providers.
- [Standard Chat WebSocket](https://docs.traceport.ai/api-reference/realtime/chat-ws.md): A simple wrapper around the Chat Completion API for persistent, multi-turn text conversations via WebSocket. Clients send standard `ChatRequest` JSON objects and receive streaming `ChatResponse` chunks.
- [Realtime API](https://docs.traceport.ai/api-reference/realtime/openai-realtime.md): Implements the OpenAI Realtime Protocol over WebSockets for ultra-low latency audio and text. Supports stateful session recovery using the `session_id` parameter.
- [Real-time & WebSockets](https://docs.traceport.ai/api-reference/realtime/overview.md): Build interactive, low-latency AI applications with persistent connections.
- [Nodes](https://docs.traceport.ai/config-workflows/nodes.md): Reference for all node types available in Config Workflows.
- [Config Workflows](https://docs.traceport.ai/config-workflows/overview.md): Build visual routing flows with plugins, routers, branches, and fallback logic.
- [Versions](https://docs.traceport.ai/config-workflows/versions.md): Manage draft, active, and inactive workflow versions.
- [What is Traceport](https://docs.traceport.ai/getting-started/what-is-traceport.md): The Unified AI Gateway — one API for all your AI providers.
- [Why Traceport](https://docs.traceport.ai/getting-started/why-traceport.md): Unified API, cost optimization, reliability, and deep observability for your AI stack.
- [Introduction](https://docs.traceport.ai/index.md): Traceport — AI observability and management platform for engineering teams.
- [Alerts and Monitoring](https://docs.traceport.ai/management/alerts.md): Monitor the health, performance, and costs of your AI applications with real-time alerts.
- [API Keys](https://docs.traceport.ai/management/api-keys.md): Manage programmatic access to the Traceport Gateway.
- [Integrations](https://docs.traceport.ai/management/integrations.md): Connect AI model providers to the Traceport Gateway.
- [Dashboard](https://docs.traceport.ai/overview/dashboard.md): Real-time analytics and metrics for your AI operations.
- [Logs](https://docs.traceport.ai/overview/logs.md): View real-time completions, token usage, latency, and costs for every LLM request.
- [Session Tracking](https://docs.traceport.ai/overview/sessions.md): Trace and audit multi-turn conversations with high-fidelity debugging tools.
- [Workflow Runs](https://docs.traceport.ai/overview/workflow-runs.md): Monitor complex, multi-step LLM workflows with execution timelines and trace debugging.
- [Editor](https://docs.traceport.ai/prompt-studio/editor.md): The prompt authoring workspace — message blocks, model parameters, and template variables.
- [Evaluations](https://docs.traceport.ai/prompt-studio/evaluations.md): Define scoring rules to automatically grade prompt quality, relevance, and safety.
- [Prompt Studio](https://docs.traceport.ai/prompt-studio/overview.md): Centralized prompt management with version control, testing, and API access.
- [Testing & Comparison](https://docs.traceport.ai/prompt-studio/testing.md): Test prompts across multiple models simultaneously and compare results side by side.
- [Quickstart](https://docs.traceport.ai/quickstart.md): Start observing and managing your LLMs with Traceport in under 5 minutes.

## OpenAPI Specs

- [apis](https://docs.traceport.ai/apis.yaml)
- [openapi](https://docs.traceport.ai/api-reference/openapi.json)

## Optional

- [Dashboard](https://app.traceport.ai)