# Pioneer > Docs for building, fine-tuning, evaluating, and deploying models with Pioneer. ## Get Started - [Drop us in. We'll ship the models.](https://docs.fastino.ai/introduction.md): Pioneer spots where your model fails, then quietly retrains it on your own data — fine-tuning, evaluation, and deployment with no MLOps team required. - [Pioneer quickstart: from signup to your first inference](https://docs.fastino.ai/quickstart.md): Go from zero to a working Pioneer inference call in minutes. Generate an API key, browse available models, and run your first NER prediction. - [How to authenticate your requests with Pioneer API](https://docs.fastino.ai/authentication.md): Generate an API key from your Pioneer account, then include it in the X-API-Key header on every request. No OAuth or token refresh required. - [Set up Claude Opus 5 with Pioneer in 60 seconds](https://docs.fastino.ai/api-reference/integrating-with-opus-5.md): Point Claude Code at Pioneer's inference endpoint and start using Claude Opus 5 in under a minute, then swap models on the fly with the /model command. ## Integrations - [Codex](https://docs.fastino.ai/codex.md): Configure the OpenAI Codex CLI to run against Pioneer's OpenAI-compatible endpoint, load a fresh model catalog, and default to pioneer/auto routing. - [Cursor](https://docs.fastino.ai/cursor.md): Configure Cursor's chat panel to use Pioneer's OpenAI-compatible API, add custom models, and switch between them from the model picker. - [OpenClaw](https://docs.fastino.ai/openclaw.md): Configure OpenClaw to use Pioneer's OpenAI-compatible endpoint, discover models via /v1/models, and run the local gateway with Pioneer Auto. - [Hermes Agent](https://docs.fastino.ai/hermes.md): Configure Hermes Agent with Pioneer using a one-time setup command that imports a filtered model catalog and defaults to Claude Opus 5. - [OpenCode](https://docs.fastino.ai/opencode.md): Connect the OpenCode CLI or desktop app to Pioneer to route model calls, switch between 70+ models, and optionally enable Exa-powered web search. ## Core Concepts - [Pioneer datasets: create, version, inspect, and delete](https://docs.fastino.ai/concepts/datasets.md): Pioneer stores and versions your training datasets automatically. Learn how to create them via generation or auto-labeling, then list, inspect, and delete them. - [Pioneer training jobs: lifecycle, metrics, and weights](https://docs.fastino.ai/concepts/training.md): Understand how Pioneer training jobs work — from submitting a job and polling status to reading metrics, stopping jobs, and downloading trained model weights. - [Inference on Pioneer: native, OpenAI, and Anthropic APIs](https://docs.fastino.ai/concepts/inference.md): Run inference on Pioneer via the native /inference endpoint, OpenAI-compatible chat completions, or Anthropic-compatible messages — all reach the same models. - [Model evaluations in Pioneer: F1, precision, recall](https://docs.fastino.ai/concepts/evaluations.md): Run Pioneer evaluations to measure F1, precision, and recall on a labeled dataset before deploying your fine-tuned model to production traffic. ### Models - [Pioneer model catalog: encoders, decoders, and inference](https://docs.fastino.ai/concepts/models.md): Browse Pioneer's encoder (GLiNER) and decoder (LLM) models for fine-tuning and inference. Covers on-demand vs. serverless and how to query the live catalog. - [GLiGuard: Safety Moderation SLM](https://docs.fastino.ai/concepts/g-li-guard.md): Run GLiGuard, Pioneer's open-source 300M safety moderation SLM, to classify prompts and completions for harmful, unsafe, or policy-violating content. - [GLiNER2-PII: PII Detection SLM](https://docs.fastino.ai/concepts/g-li-ner-2-pii.md): Run GLiNER2-PII, Pioneer's open-source multilingual PII detection SLM, to identify names, emails, phone numbers, and other personal data in text. ## Guides ### Fine-tuning - [Fine-tune Nemotron 3.5 Lightning on Pioneer](https://docs.fastino.ai/guides/fine-tune-llm.md): LoRA fine-tune Nemotron 3.5 Lightning on Pioneer with supervised fine-tuning via one training endpoint — from dataset prep to a deployed decoder model. #### Encoder fine-tuning - [Fine-tune a GLiNER NER model from data to inference](https://docs.fastino.ai/guides/fine-tune-ner.md): Train a custom Named Entity Recognition model on your data using Pioneer's GLiNER encoder models, from dataset prep through evaluation and inference. - [Fine-tune a GLiNER text classification model on Pioneer](https://docs.fastino.ai/guides/fine-tune-classification.md): Train a custom single- or multi-label text classification model on Pioneer's GLiNER encoders, from dataset prep through evaluation and inference on your data. - [Fine-tune a GLiNER structured extraction model on Pioneer](https://docs.fastino.ai/guides/fine-tune-extraction.md): Train a custom JSON extraction model on Pioneer's GLiNER encoders to pull invoices, forms, and structured records out of unstructured text — data to inference. ### Data & Agents - [Generate synthetic training data for NER and LLM tasks](https://docs.fastino.ai/guides/synthetic-data.md): Use Pioneer's data generation API to create labeled NER, classification, and decoder training examples without manual annotation, or auto-label existing text. - [Adaptive Inference: automatic continuous retraining](https://docs.fastino.ai/guides/adaptive-inference.md): Pioneer's Adaptive Inference monitors live traffic, collects corrections, retrains a new checkpoint, and promotes it automatically when performance improves. - [Use Pioneer with AI coding agents via Agent Skills](https://docs.fastino.ai/guides/agent-skills.md): Add a SKILL.md file to your AI coding agent so Cursor, Claude Code, or similar agents can manage Pioneer datasets, training, and inference autonomously. ## Account - [Trust & Safety](https://docs.fastino.ai/trust-safety.md): How Pioneer protects your data, secures its infrastructure, and builds AI you can deploy responsibly. ## Other - [AGENTS](https://docs.fastino.ai/AGENTS.md) - [Pioneer projects API: deploy and manage model endpoints](https://docs.fastino.ai/api-reference/projects.md): Organize Pioneer models into projects, deploy fine-tuned models to project endpoints, and run inference without managing training job IDs in your application. - [Claude Code](https://docs.fastino.ai/claude-code.md): Point Claude Code at Pioneer for multi-model inference with router-backed pioneer/auto routing, model discovery, and full /model picker support inside the CLI. - [Pioneer: fine-tune and deploy open-source AI models](https://docs.fastino.ai/index.md): Pioneer is an AI fine-tuning and inference platform. Train custom LLMs and NER models, run inference, and deploy — all through a simple REST API. ## OpenAPI Specs - [openapi](/openapi.json) > The links below point to documentation indexes. Follow each `/_llms/` index recursively until you reach documentation pages. ## Indexes - [Chinese (13 pages)](https://docs.fastino.ai/_llms/cn.md): Documentation for Chinese. - [Spanish (13 pages)](https://docs.fastino.ai/_llms/es.md): Documentation for Spanish. - [French (13 pages)](https://docs.fastino.ai/_llms/fr.md): Documentation for French. - [German (13 pages)](https://docs.fastino.ai/_llms/de.md): Documentation for German.