GDP.pdf leaderboard
As of November 14, 2026, among the results listed here, Sakana Fugu Ultra v2 has the highest reported GDP.pdf score, 34.3%, published by Sakana, ahead of Meta Muse Wonder at 30.8%. 9 models have a GDP.pdf result in our records.
Scores as the model makers published them, not re-run by us; each row links to its announcement.
9 models ranked · 5 makers
| Rank | Model | Score | Setting | Reported by | Date |
|---|---|---|---|---|---|
| 1 | Fugu Ultra v2 | 34.3% | Not published | Sakana | June 24, 2026 (model release) · read from the maker's chart |
| 2 | Muse Wonder | 30.8% | max | Meta | October 12, 2026 (model release) |
| 3 | GPT-5.6 Sol | 30.7% | Not published | OpenAI | August 7, 2025 (model release) |
| 4 | GPT-5.5 | 26.0% | Not published | OpenAI | August 7, 2025 (model release) |
| 4 | Fugu Max | 26.0% | Not published | Sakana | June 24, 2026 (model release) · read from the maker's chart |
| 6 | T Inkling 2 Small | 25.2% | Not published | Thinking Machines | November 9, 2026 (model release) |
| 7 | GPT-5.6 Terra | 24.7% | Not published | OpenAI | August 7, 2025 (model release) |
| 8 | GPT-5.6 Luna | 22.7% | Not published | OpenAI | August 7, 2025 (model release) |
| 9 | Step 5 Preview | 14.8% | High | StepFun | October 8, 2026 (model release) |
How much to trust these numbers
- Source
- Reported by the model makers themselves. Not re-run by LLMBenchmark or anyone independent.
- Settings
- Settings differ between makers. Check the Setting column before comparing two scores.
- Spread
- 34.3% to 14.8% across 9 models.
- Freshness
- Newest result published November 9, 2026; oldest August 7, 2025.
- Charts
- Some values were read from charts in the announcements. Two independent readings had to match exactly.
What GDP.pdf measures
GDP.pdf asks models to reason over professional PDFs using real-world prompts; the search results do not specify what its numerical score means. Source: GDP.pdf: Benchmarking Grounded Multimodal Reasoning over Professional PDF Documents