LLM Stats (GDP.pdf): leaderboard
LLM Stats mirror of GDP.pdf — multimodal models tackle economically valuable professional tasks presented as PDF documents, testing document-based reasoning and chart/table interpretation without tools.
Metric: Score (%). Source: llm-stats.com. Saturation forecast: Around 2029. 12 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Muse Wonder Full | 33.2 | #10 |
| 2 | GPT-6.1 Sol | 32 | #7 |
| 3 | Muse Wonder | 30.8 | #15 |
| 4 | GPT-5.6 Sol | 30.7 | #19 |
| 5 | Claude Fable 5 | 29.8 | #16 |
| 6 | GPT-5.6 Terra | 24.7 | #41 |
| 7 | GPT-5.6 Luna | 22.7 | #68 |
| 8 | MiniMax-M3.1-Flash | 20.8 | #33 |
| 9 | Mistral Large 4 | 18.6 | #94 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #8 Gemini 4 Argon.
Interactive version: theaggregate.ai/benchmark?slug=llm-stats-gdp-pdf · How It Works · Data refreshed daily, snapshot 2026-10-20.