# GDPval-AA — AI model rankings

Measures how well the AI does economically valuable knowledge work, judged against human experts. Shown as a rating (like a chess Elo) — higher is better.

9 tracked models have a published GDPval-AA score. Higher is better. Scores are as published at each model's release.

## Ranking

| Rank | Model | Developer | Score | Released |
| --- | --- | --- | --- | --- |
| 1 | Claude Fable 5 | Anthropic | 1932 | Jun 9 2026 |
| 2 | Claude Opus 4.8 | Anthropic | 1890 | May 28 2026 |
| 3 | GPT-5.5 | OpenAI | 1769 | Apr 23 2026 |
| 4 | Claude Opus 4.7 | Anthropic | 1753 | Apr 16 2026 |
| 5 | Claude Sonnet 4.6 | Anthropic | 1676 | Feb 17 2026 |
| 6 | Gemini 3.5 Flash | Google | 1656 | May 19 2026 |
| 7 | Claude Sonnet 5 | Anthropic | 1618 | Jun 30 2026 |
| 8 | Gemini 3.1 Pro | Google | 1314 | Feb 19 2026 |
| 9 | Gemini 3.0 Flash | Google | 1204 | Dec 17 2025 |


---

Canonical page: https://aireleasetracker.com/benchmark/gdpval-aa
Full dataset: https://aireleasetracker.com/llms-full.txt · JSON: https://aireleasetracker.com/models.json
Source: AI Release Tracker (https://aireleasetracker.com). Benchmark scores are the figures published by the releasing lab at launch.