AI is changing how work gets done. Explore the main ways to use a language model, from a single chat message to teams of agents that run on their own, grouped into eight levels by how much the model decides for itself.
No background needed. See how a chatbot grows into an agent, then find the level that fits your own work.
Name the kind of job, pick the smallest level that does it, and give your own model a brief it can act on.
See how conversational AI expanded through stronger reasoning, retrieval, workflows, tools, agent loops, and systems that keep working across sessions.
Overlapping developments, explained through seven levels. Teams are optional.
Find a concept, trace its prerequisites, and see what it can lead to. The full map connects 54 concepts across the eight levels and the topics that run through them.
Explore the mapWant to see a concept in practice? Explore worked examples.
One task, two designs. Compare coverage, effort, and reliability; choose the design that fits the task.
The question: “How long is the warranty on the DW-480, and what voids it?”
Your code sets the search and passes the retrieved passages to the model.
The answer gives a two-year warranty and two exclusions. A third exclusion is missing from the retrieved excerpts.
A predictable sequence, but its answer depends on the evidence retrieved. RAG can also use richer retrieval and validation; one search is this design’s choice.
Explore the RAG guided example →The model chooses follow-up searches and which documents to open, within tool permissions and run limits.
Opening the full exclusions section reveals a third condition: service by an unlisted technician also voids the warranty.
More chances to fill a gap, with extra calls and opportunities for error. The model can still stop too early; citations and coverage need checking.
Explore the agentic RAG guided example →Fictional, scripted comparison—not a benchmark or a claim about every document-chat product. The linked examples explore these techniques in other scenarios.
The timeline
Selected publications and public milestones, not dates of invention. Select a product for the event, date and evidence.
0 · Conventional software Rules, search and conventional software have no single arrival date.
Lines connect selected dates, not development time. Sources checked through Sep 20, 2026.
Models trained to think before they answer. Not a new level: a better engine for every level, and the one OpenAI credits for its first agent. Source ↗
TypeSafe introduced Jev in early access on September 15, 2026: a model that returns typed decisions with probabilities rather than generated text. “System One” is TypeSafe’s name for its proposed model category. Source ↗
An example of agents that keep working beyond an interactive session. The product also describes a team of agents: one product can illustrate more than one level.
Marked event: Aug 11, 2026 · Introducing Grok Bot ↗
Foundational publication: Generative Agents ↗ · Apr 7, 2023. This is a reference point, not a claim to the first use of the idea.
Anthropic explicitly describes a lead agent delegating to parallel subagents on June 13. Research launched on April 15, but that announcement does not establish whether the same architecture was in use then. June is the documentation date, not an asserted deployment date.
Marked event: Jun 13, 2025 · How we built our multi-agent research system ↗
Foundational publication: CAMEL ↗ · Mar 31, 2023. This is a reference point, not a claim to the first use of the idea.
The agent observes a browser, acts, and decides what to do next. This is a browser-use milestone; agents and autonomous loops existed earlier.
Marked event: Jan 23, 2025 · Introducing Operator ↗
Foundational publication: ReAct ↗ · Oct 6, 2022. This is a reference point, not a claim to the first use of the idea.
A visible example of a model choosing a tool. This dates the limited rollout, not broad availability or the first implementation of tool use.
Marked event: Mar 23, 2023 · ChatGPT plugins ↗
Foundational publication: MRKL Systems ↗ · May 1, 2022. This is a reference point, not a claim to the first use of the idea.
Later access: rollout to Plus subscribers begins, May 12, 2023.
A model call inside a workflow whose steps are set in advance. December 9 is an archived-page capture, not a confirmed launch date. Microsoft 365 Copilot remains a related workplace-AI milestone, rather than the defining workflow example.
Marked event: Available by Dec 9, 2022 · OpenAI Integrations ↗
Foundational publication: AI Chains ↗ · Oct 4, 2021. This is a reference point, not a claim to the first use of the idea.
The early Perplexity Ask combined retrieved search results with a generated answer and citations. That is the specific Level 2 pattern illustrated here, not a classification of every later Perplexity feature. Cofounder Aravind Srinivas dates the launch to December 7, 2022 in a retrospective interview; this is founder testimony, not a contemporaneous launch announcement.
Marked event: Dec 7, 2022 · Aravind Srinivas on the launch date · Stripe Sessions 2024 ↗
Early behavior: Founder interview describing the first Perplexity Ask ↗.
Foundational publication: Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks ↗ · May 22, 2020. This is a reference point, not a claim to the first use of the idea.
A recognizable public example of prompting a language model. Earlier products existed; this is a selected milestone, not the invention of prompting.
Marked event: Nov 30, 2022 · ChatGPT ↗
Foundational publication: Better Language Models and Their Implications (GPT-2) ↗ · Feb 14, 2019. This is a reference point, not a claim to the first use of the idea.
Epoch AI’s Epoch Capabilities Index, plotted by model release date, with the best score to date and this guide’s level milestones.
Anthropic · ECI 163.27 · Highest score Jun 9, 2026–Aug 31, 2026
Epoch AI’s Capabilities Index combines scores from many different AI benchmarks into a single “general capability” scale, allowing comparisons between models even over timespans long enough for single benchmarks to reach saturation. Epoch adds that “Absolute ECI values are meaningless by themselves, but meaningful comparisons can be made between models.” So read the shape, not the numbers. Record periods are reconstructed from this snapshot’s scores and release dates, not historical leaderboard snapshots. Highest ECI does not mean best at every task.
The shape has a bend in it. From GPT-4 on Mar 14, 2023 to the day before the first reasoning model, 18 months, the best score rose 4.1 points, from 125.89 to 130. In the 24 months since, it has risen 36.3, to 166.31 (GPT-6 Astra, Sep 3, 2026). The selected milestones for levels 1 to 4 fall before the bend; the selected milestones for levels 5 to 7 fall after it. This reflects the editorial selection, not the first emergence of those patterns. Both figures are this site’s subtraction over Epoch’s scores. The index starts in early 2023, so it says nothing about what came before.
| Released | Model | Maker | Score |
|---|---|---|---|
| 02/24/2023 | LLaMA-65B | Meta AI | 109.94 |
| 03/14/2023 | GPT-4 (Mar 2023) | OpenAI | 125.89 |
| 01/25/2024 | GPT-4 Turbo (Nov 2023) | OpenAI | 126.45 |
| 02/29/2024 | Claude 3 Opus | Anthropic | 126.91 |
| 04/09/2024 | GPT-4 Turbo (Apr 2024) | OpenAI | 127.25 |
| 05/13/2024 | GPT-4o (May 2024) | OpenAI | 128.97 |
| 06/20/2024 | Claude 3.5 Sonnet | Anthropic | 130 |
| 09/12/2024 | o1-mini | OpenAI | 135.82 |
| 12/17/2024 | o1 | OpenAI | 141.86 |
| 03/31/2025 | Gemini 2.5 Pro (Mar 2025) | Google DeepMind | 144.16 |
| 04/16/2025 | o3 | OpenAI | 146.91 |
| 06/10/2025 | o3-pro | OpenAI | 147.45 |
| 08/07/2025 | GPT-5 | OpenAI | 150 |
| 10/07/2025 | GPT-5 Pro | OpenAI | 150.29 |
| 11/18/2025 | Gemini 3 Pro | Google DeepMind | 153 |
| 12/11/2025 | GPT-5.2 Pro | OpenAI | 155.38 |
| 02/05/2026 | GPT-5.3 Codex | OpenAI | 156.58 |
| 03/05/2026 | GPT-5.4 Pro | OpenAI | 158.95 |
| 04/23/2026 | GPT-5.5 Pro | OpenAI | 162.25 |
| 06/09/2026 | Claude Fable 5 | Anthropic | 163.27 |
| 09/01/2026 | Claude Fable 5.1 | Anthropic | 164.47 |
| 09/03/2026 | GPT-6 Astra | OpenAI | 166.31 |
Data: Epoch Capabilities Index, Epoch AI, reused under CC BY 4.0, retrieved 09/18/2026; 266 models. Scores, release dates and makers are Epoch’s, unchanged apart from rounding. The level markers and the reasoning line are this site’s. Best score on the day of the snapshot: 166.31.
Our eight levels organize patterns by the decisions a model can make. Higher does not mean better, and a system can combine several patterns. Start with your task, then explore the concepts it needs.
Describe your task, prepare a brief, and give it to your own model with representative files. The model can consult this guide and recommend an approach; the site itself does not run it.