# GDP.PDF — AI model rankings

Real professional PDFs — filings, reports, technical documents — with questions an expert in that field would ask. Tests whether the AI reads the page as a document, layout and figures included, rather than as loose text. Higher is better.

10 tracked models have a published GDP.PDF score. Higher is better. Scores come from published lab reports and benchmark sources; recorded sources appear beside the scores.

## Ranking

| Rank | Model | Developer | Score | Source | Released |
| --- | --- | --- | --- | --- | --- |
| 1 | Gemini 3.8 Flash | Google | 35% | Lab | Sep 2 2026 |
| 2 | Kimi K3.1 Pro | Moonshot AI | 34.6% | Lab | Oct 23 2026 |
| 3 | Gemini 3.7 Flash | Google | 34% | Lab | Aug 13 2026 |
| 4 | Muse Wonder Full | Meta | 33.2% | Lab | Oct 12 2026 |
| 5 | GPT-6.1 Sol | OpenAI | 32% | Lab | Sep 29 2026 |
| 6 | Muse Wonder | Meta | 30.8% | Lab | Oct 12 2026 |
| 7 | GPT-5.6 Sol | OpenAI | 30.7% | Lab | Jun 26 2026 |
| 8 | GPT-5.6 Terra | OpenAI | 24.7% | Lab | Jun 26 2026 |
| 9 | GPT-5.6 Luna | OpenAI | 22.7% | Lab | Jun 26 2026 |
| 10 | Gemini 3.6 Flash | Google | 22% | Lab | Jul 21 2026 |


---

Canonical page: https://aireleasetracker.com/benchmark/gdp-pdf
Site index: https://aireleasetracker.com/llms.txt
Source: AI Release Tracker (https://aireleasetracker.com). Most benchmark scores come from lab launch material; gathered results identify the leaderboard that published them.