---
title: Pictograph
description: Build computer-vision datasets and train models from Python, the CLI, or an agent.
section: Get Started
order: -1
---
Pictograph turns directories of images into trained CV models with as little hand-annotation as possible. The same REST API drives three surfaces: a typed Python SDK, a CLI, and an agent toolkit for Claude and OpenAI.

```python
from pictograph import Client
from pictograph.workflows import full_pipeline

client = Client()
report = full_pipeline(
    client,
    dataset_name="road-signs",
    folder="./road_signs",
    classes=[("stop_sign", "bbox"), ("yield", "bbox")],
    pipeline="yolox",
)
print("model:", report.model.id if report.success else "see report")
```

## What you can do

- **Upload** directories of images; subdirectories become virtual paths.
- **Auto-annotate** with SAM3 — point, box, or text prompts, single image or async batch.
- **Train** YOLOX, Detectron2, SM-PyTorch, RF-DETR, or classification models on A10G / A100 / H100 GPUs.
- **Export** to COCO, YOLO, CVAT, Pascal VOC, LabelMe, CSV, or Pictograph JSON.
- **Import** existing datasets from V7 (Darwin) or Roboflow.
- **Search** by visual similarity (SigLIP2) or auto-generated content tags.
- **Drive everything from agents** — Claude Agent SDK, openai-agents, Vercel AI SDK, LangChain, or any framework that speaks JSON Schema.

## Map of the docs

| Section | Pages |
| --- | --- |
| **Get Started** | [Installation](/docs/installation.md) · [Quickstart](/docs/quick-start.md) · [Authentication](/docs/authentication.md) |
| **Workflows** | [Full pipeline](/docs/workflows/full-pipeline.md) · [Upload](/docs/workflows/upload.md) · [Auto-annotate](/docs/workflows/auto-annotate.md) · [Train](/docs/workflows/train.md) |
| **API Reference** | [Overview](/docs/api-reference.md) · [Datasets](/docs/api-reference/datasets.md) · [Images](/docs/api-reference/images.md) · [Annotations](/docs/api-reference/annotations.md) · [Auto-annotate](/docs/api-reference/auto-annotate.md) · [Search](/docs/api-reference/search.md) · [Batch](/docs/api-reference/batch.md) · [Exports](/docs/api-reference/exports.md) · [Training](/docs/api-reference/training.md) · [Models](/docs/api-reference/models.md) · [Credits](/docs/api-reference/credits.md) · [Connectors](/docs/api-reference/connectors.md) · [Video](/docs/api-reference/video.md) · [Organizations](/docs/api-reference/organizations.md) · [Projects](/docs/api-reference/projects.md) · [API Keys](/docs/api-reference/api-keys.md) · [Tools](/docs/api-reference/tools.md) |
| **Agents** | [Overview](/docs/agents.md) · [Claude](/docs/agents/claude.md) · [OpenAI](/docs/agents/openai.md) · [Dynamic discovery](/docs/agents/dynamic-discovery.md) · [Cookbook](/docs/agents/cookbook.md) |
| **Reference** | [Annotation format](/docs/annotation-format.md) · [Error handling](/docs/error-handling.md) · [Rate limits](/docs/rate-limits.md) · [CLI](/docs/cli.md) |

Every page has a "Copy as Markdown" button and an `.md` mirror for agents to consume directly.

## For agents browsing this site

- Site index: [`/docs/llms.txt`](/docs/llms.txt)
- Full doc bundle (one file): [`/docs/llms-full.txt`](/docs/llms-full.txt)
- Tool registry (JSON Schema): [`/api/v1/developer/tools.json`](https://api.pictograph.io/api/v1/developer/tools.json)