# Adaption The full text of every page below is in one file: [llms-full.txt](/llms-full.txt) ## Docs - [Adaptive Data quickstart](/adaptive-data-quickstart/index.md): Install the Adaption Python SDK, import data, run Adaptive Data, and download the adapted dataset. - [Combine datasets](/adaptive-data/combine-datasets/index.md): Validate and merge multiple Adaptive Data datasets into one new dataset with the Python SDK. - [Configure Adaptive Data](/adaptive-data/configure-adaptive-data/index.md): Generate instruction or preference data, apply recipes, limit rows, and control adapted outputs. - [Create a dataset](/adaptive-data/create-a-dataset/index.md): Upload a local file or import a dataset from Hugging Face or Kaggle with the Python SDK. - [Evaluate dataset quality](/adaptive-data/evaluate-dataset-quality/index.md): Poll evaluation status and inspect quality metrics for an adapted dataset with the Python SDK. - [Expand your data](/adaptive-data/expand-data/index.md): Augment, generate similar, translate, or localize rows to increase dataset size after import or adaptation. - [Extend your trajectories](/adaptive-data/extend-trajectories/index.md): Generate a next user turn for each conversation to build multi-turn training data that depends on earlier context. - [Invent a dataset](/adaptive-data/invent-a-dataset/index.md): Generate a model-ready dataset from scratch, with no source dataset required. - [Overview](/adaptive-data/overview/index.md): Understand what Adaptive Data produces and how data moves from import through adaptation and evaluation. - [Select columns to adapt](/adaptive-data/select-columns/index.md): Map dataset columns to prompt, completion, context, image, or chat roles before running Adaptive Data. - [AutoScientist quickstart](/autoscientist-quickstart/index.md): Create a training-ready dataset, run AutoScientist, and download the best model checkpoint. - [Data augmentation](/autoscientist/data-augmentation/index.md): Add domain-specific and general-purpose rows before AutoScientist training begins. - [Download the model](/autoscientist/download-the-model/index.md): Stream the best AutoScientist checkpoint to a local archive. - [Interpreting results](/autoscientist/interpreting-results/index.md): Retrieve an AutoScientist run and understand its win rate and training diagnostics. - [Overview](/autoscientist/overview/index.md): Understand what AutoScientist builds and how its iterative training loop works. - [Hyperparameters](/autoscientist/recommended-hyperparameters/index.md): Understand the training recipe AutoScientist selects and iterates on automatically. - [Run on non-adapted data](/autoscientist/run-on-non-adapted-data/index.md): Upload training-ready prompt/completion data for supervised fine-tuning. - [Running AutoScientist](/autoscientist/running-autoscientist/index.md): Create, monitor, and handle an AutoScientist training run with the Python SDK. - [Supported models](/autoscientist/supported-models/index.md): Base models available to AutoScientist and the per-model limits for context length, batch size, and LoRA rank. - [Multilingual datasets](/guides/multilingual-datasets/index.md): Multilingual adaptation across the data space is core to Adaptive Data; dedicated SDK configuration is coming soon. - [Adaption documentation](/index.md): Start with Adaptive Data, AutoScientist, and the Adaption API. - [Generate a model response](/inference-stacks/generate-a-model-response/index.md): Choose a model, continue conversations, stream output, use tools, and send image inputs with the Responses API. - [Key concepts](/inference-stacks/key-concepts/index.md): Understand stacks, tools and connectors, file search, and memory in the Adaption Responses API. - [Models](/inference-stacks/models/index.md): Compare pricing, context lengths, and input modalities for models available through the Adaption Responses API. - [Quickstart](/inference-stacks/quickstart/index.md): Create an Adaption API key, install the Python SDK, and generate your first model response. - [Create API keys](/introduction/create-api-keys/index.md): Create and use Adaption API keys for secure programmatic access to Adaptive Data from applications, CI, and notebooks. - [Getting started](/introduction/getting-started/index.md): Install the Adaption Python SDK and authenticate with an API key. - [OAuth](/oauth/index.md): How MCP clients sign in to the Adaption MCP server with OAuth 2.1, and what the authorization server supports. - [FAQ](/resources/faq/index.md): Frequently asked questions about Adaptive Data, the Adaption API, and the Python SDK. - [Support](/resources/support/index.md): Get product and API help from the Adaption team and community on Discord. - [Adaptive Data App Walkthrough](/runbooks/adaptive-data-app-walkthrough/index.md): Import a dataset, map its columns, configure adaptation, launch a run, and review the results in the Adaption app. - [AutoScientist App Walkthrough](/runbooks/autoscientist-app-walkthrough/index.md): Select training data, review AutoScientist's proposed model and recipe, launch training, and interpret the result in the Adaption app. - [Expand a Dataset](/runbooks/expand-a-dataset/index.md): Upload 500 GSM8K rows, grow them with domain, general, and similar rows, and compare what each kind of row looks like. - [Extend Conversation Trajectories](/runbooks/extend-trajectories/index.md): Generate next user turns for WildChat conversations that a model can only answer well by using the whole conversation. - [Invent a Dataset](/runbooks/invent-a-dataset/index.md): Generate a 100-row medical Q&A dataset from a single prompt, download it, and score its diversity with DCScore. - [Overview](/runbooks/overview/index.md): Step-by-step workflows for the Adaption API and app. Run the notebooks in Google Colab or follow along in the docs. - [Processing Unstructured Documents](/runbooks/processing-unstructured-documents/index.md): Ingest PDFs, documents, spreadsheets, and slide decks in the Adaptive Data app without building a preprocessing pipeline. - [Train a Model with AutoScientist](/runbooks/train-with-autoscientist/index.md): Import the Alpaca dataset from Hugging Face, train with AutoScientist and data augmentation, and download the best checkpoint. - [Claude plugin](/tools/claude-plugin/index.md): Connect Claude to Adaption with MCP tools and skills for dataset and model-training workflows in chat, Cowork, and Claude Code. - [ChatGPT and Codex plugin](/tools/codex-plugin/index.md): Connect ChatGPT and Codex to Adaption with MCP tools and skills for dataset and model-training workflows. - [Cursor plugin](/tools/cursor-plugin/index.md): Connect Cursor to Adaption with MCP tools and skills for dataset and model-training workflows. - [Hermes plugin](/tools/hermes-plugin/index.md): Connect Hermes Agent to Adaption with the hosted MCP server and skills for dataset and model-training workflows. - [MCP](/tools/mcp/index.md): Connect an MCP client to Adaption and use tools for datasets, synthetic data generation, and model training. - [OpenCode](/tools/opencode/index.md): Configure Adaption as an inference provider in OpenCode and code with Adaption models. ## API Reference Links below point to language-neutral HTTP documentation. SDK-specific docs follow the same URL structure with the language inserted after the base path: `/api/{language}/resources/...`. Available languages: http, python ### [Datasets](/api/resources/datasets/index.md) - [Create](/api/resources/datasets/methods/create/index.md): Create a dataset from file upload, HuggingFace, Kaggle, Google Sheets, or a public S3 or Cloud Storage object - [Get](/api/resources/datasets/methods/get/index.md): Get a dataset by ID - [List](/api/resources/datasets/methods/list/index.md): List datasets - [Get Status](/api/resources/datasets/methods/get_status/index.md): Get the processing status of a dataset - [Download](/api/resources/datasets/methods/download/index.md): Download the processed dataset - [Publish](/api/resources/datasets/methods/publish/index.md): Publish a dataset to an external platform - [Run](/api/resources/datasets/methods/run/index.md): Adapt a dataset (deprecated) - [Adapt](/api/resources/datasets/methods/adapt/index.md): Adapt a dataset (or estimate cost) - [Augment](/api/resources/datasets/methods/augment/index.md): Add rows to a dataset from the curated pool - [Translate](/api/resources/datasets/methods/translate/index.md): Add translated rows to a dataset - [Localize](/api/resources/datasets/methods/localize/index.md): Add localized rows to a dataset - [Extend Trajectories](/api/resources/datasets/methods/extend_trajectories/index.md): Extend trajectories with a next user turn (or estimate cost) - [Get Evaluation](/api/resources/datasets/methods/get_evaluation/index.md): Get evaluation results for a dataset - [Invent](/api/resources/datasets/methods/invent/index.md): Generate a dataset from scratch - [Invent Domains](/api/resources/datasets/methods/invent_domains/index.md): List the domains and subdomains available for generation - [Delete](/api/resources/datasets/methods/delete/index.md): Delete a dataset - [Get Export Links](/api/resources/datasets/methods/get_export_links/index.md): Poll saved URLs for successful dataset exports (Hugging Face, Kaggle, Google Sheets) - [Get Best Launch Config](/api/resources/datasets/methods/get_best_launch_config/index.md): Best fine-tune launch config for re-launch parity - [Get Preprocessing Credits](/api/resources/datasets/methods/get_preprocessing_credits/index.md): Estimate credit cost for document preprocessing - [Launch Preprocessing](/api/resources/datasets/methods/launch_preprocessing/index.md): Launch document preprocessing - [Upload](/api/resources/datasets/subresources/upload/index.md) - [Initiate](/api/resources/datasets/subresources/upload/methods/initiate/index.md): Initiate a dataset upload - [Complete](/api/resources/datasets/subresources/upload/methods/complete/index.md): Complete a dataset upload and trigger processing - [Complete By ID](/api/resources/datasets/subresources/upload/methods/complete_by_id/index.md): Complete a file upload and trigger processing - [Initiate Batch](/api/resources/datasets/subresources/upload/methods/initiate_batch/index.md): Initiate a batch upload - [Complete Batch](/api/resources/datasets/subresources/upload/methods/complete_batch/index.md): Complete a batch upload and trigger processing - [Combine](/api/resources/datasets/subresources/combine/index.md) - [Create](/api/resources/datasets/subresources/combine/methods/create/index.md): Combine multiple Datasets into a single new merged Dataset - [Validate](/api/resources/datasets/subresources/combine/methods/validate/index.md): Validate that a set of Datasets can be combined - [Custom Evals](/api/resources/datasets/subresources/custom_evals/index.md) - [Create](/api/resources/datasets/subresources/custom_evals/methods/create/index.md): Start a custom-rubric evaluation for an adapted dataset - [List](/api/resources/datasets/subresources/custom_evals/methods/list/index.md): List custom-rubric evaluations for a dataset, newest first - [Get](/api/resources/datasets/subresources/custom_evals/methods/get/index.md): Get a single custom-rubric evaluation ### [Autoscientist](/api/resources/autoscientist/index.md) - [Create](/api/resources/autoscientist/methods/create/index.md): Create an AutoScientist run - [Recommend Hyperparams](/api/resources/autoscientist/methods/recommend_hyperparams/index.md): Recommend hyperparameters - [Get](/api/resources/autoscientist/methods/get/index.md): Retrieve an AutoScientist run - [List](/api/resources/autoscientist/methods/list/index.md): List AutoScientist runs - [Cancel](/api/resources/autoscientist/methods/cancel/index.md): Cancel an AutoScientist run - [Download](/api/resources/autoscientist/methods/download/index.md): Download the best trained model artifact - [List Models](/api/resources/autoscientist/methods/list_models/index.md): List available base models for training ### [Training Models](/api/resources/training_models/index.md) - [List](/api/resources/training_models/methods/list/index.md): List available base models for training (deprecated) ### [Responses](/api/resources/responses/index.md) - [Create](/api/resources/responses/methods/create/index.md): Generate a model response - [Retrieve](/api/resources/responses/methods/retrieve/index.md): Retrieve a model response - [Delete](/api/resources/responses/methods/delete/index.md): Delete a model response - [Input Items](/api/resources/responses/subresources/input_items/index.md) - [List](/api/resources/responses/subresources/input_items/methods/list/index.md): List response input items ### [Chat](/api/resources/chat/index.md) - [Completions](/api/resources/chat/subresources/completions/index.md) - [Create](/api/resources/chat/subresources/completions/methods/create/index.md): Create a chat completion ### [Tools](/api/resources/tools/index.md) - [List](/api/resources/tools/methods/list/index.md): List host-side tools ### [Models](/api/resources/models/index.md) - [List](/api/resources/models/methods/list/index.md): List supported models