LANGUAGE MODEL Nov 2022 OpenAI

ChatGPT

RLHF-Aligned Conversational LLM

A version of GPT-3.5 fine-tuned with human feedback so that it follows instructions, holds a multi-turn conversation, and refuses obviously harmful requests. Free public access through a chat interface made it the fastest-growing consumer product in history, crossing 100 million users in two months.

Try ChatGPT API Docs ↗

Intelligence

Medium

Speed

Good

157 tok/s output

Cost

High

$1.25 in / $10.00 out

Context

Up to 4,096 tokens

How are Intelligence, Speed & Cost bucketed?

Intelligence and Speed buckets are percentile ranks on Artificial Analysis. Cost buckets are fixed dollar thresholds keyed off output-token price ($/M out).

Intelligence

Top 1%≤ 1%
Top 5%≤ 5%
Top 10%≤ 10%
Good≤ 25%
Medium≤ 50%
Below avg> 50%

Speed

Top 1%≥ 345 tok/s
Top 5%≥ 237 tok/s
Top 10%≥ 196 tok/s
Good≥ 146 tok/s
Medium≥ 90 tok/s
Slow< 90 tok/s

Cost

Freeopen weights · self-host
Low< $1 / M out
Moderate$1–5 / M out
High≥ $5 / M out

Official ↗ Artificial Analysis ↗

Why it matters

Even readers with no technical interest in AI now interact with GPT-class systems weekly — through ChatGPT, Copilot, or one of dozens of derivative products. ChatGPT is the moment AI stopped being a research field and became a consumer category. Every subsequent regulatory, geopolitical, labor-market, and education debate about AI is downstream of this launch.

Core Capabilities

Long Documents

Handles entire codebases, books, and multi-doc RAG.

Generative

Produces images, video, audio, or other media.

Research

Foundational paper or scientific contribution.

Context Window

4k tokens

≈ short doc

4k This model 本模型

32k Long docs 长文档

128k Books 整本书

400k Multi-doc 多文档

1M Codebase 整个代码库

10M

Availability

API

Not available

Product / App

Available

Open Source

Not released

Enterprise

Contact sales

Pricing Model

Subscription

Bundled inside the host product.

Subscription

Capability / Performance

Where this model sits relative to the middle 60% of models in the tree. All scores are 0–10 (higher is better).

Lower 20% Upper 80% This model

Reasoning

AA Intelligence Index · scaled to 10

1.7

5.6

3.1

Coding

SciCode · scaled to 10

1.8

4.3

3.8

Agentic tasks

Terminal-Bench Hard · scaled to 10

0.2

3.6

1.3

Context / memory

Context window size · log-scaled

6.0

9.0

1.0

Cost efficiency

Input price ($/M tokens) · cheaper scores higher

6.2

10.0

7.5

Lower 20% 20th percentile — 20% of models score below this This model Where the current model lands Upper 80% 80th percentile — only 20% of models score above this Percentile boundaries are computed across every model in the tree that reports the underlying benchmark for each capability.

What it feels like

Reached 1 million users in 5 days — fastest consumer tech adoption recorded at the time
First chat interface that made LLMs useful to non-technical users — the productisation moment
Built on InstructGPT (GPT-3.5 + RLHF), wrapped in dialogue tuning + system-prompt scaffolding
Triggered the entire AI-product wave — every major lab released a competing chatbot within 6 months
Free at launch (Nov 2022); ChatGPT Plus ($20/mo) followed Feb 2023; by 2024 ~200M WAU
Today's ChatGPT product surface is GPT-5 / GPT-4o under the hood — the brand outlived the original model

Reviews: OpenAI — Introducing ChatGPT ↗ · GPT-4 Technical Report (arXiv) ↗ · MIT Press — How is ChatGPT's behavior changing ↗

Best use cases

Casual chat, writing assistance, brainstorming, and general-purpose Q&A
Onboarding non-technical users to AI
Studying the productisation of AI as a tech-history case study

Tools to try

ChatGPT Codex CLI Cursor GitHub Copilot Continue.dev

Not ideal for

Discussions of the November 2022 model itself — the production model has changed many times
Self-hosted / regulated workloads (it's a hosted consumer / enterprise product)

Model Evolution

View full evolution tree →