GDP.pdf (Artificial Analysis, All-pass)
Professional Work
Artificial Analysis' independent GDP.pdf implementation: 100 professional document tasks, five attempts per task, LiteParse text plus page images where supported, and GPT-5.6 Luna Medium judging. All-pass credits an attempt only if every atomic criterion passes; this harness differs from Surge's implementation.
Top models (higher is better)
| Model | Score |
|---|---|
| Muse Wonder Full | 32.6 |
| GPT-6 Astra | 32.2 |
| GPT-6.1 Sol | 32.0 |
| Kimi K3.1 Pro | 31.4 |
| Muse Wonder | 30.6 |
| Claude Opus 5.5 | 28.8 |
| GPT-6.1 Terra | 28.6 |
| Gemini 4 Carbon | 28.4 |
| Claude Fable 5.1 | 28.0 |
| GPT-5.6 Sol | 27.8 |
| Muse Spark 1.3 | 26.6 |
| Claude Sonnet 5.5 | 25.8 |
| GPT-6 Sol | 25.2 |
| GPT-5.6 Terra | 24.6 |
| GPT-5.6 Luna | 24.0 |
| Gemini 3.7 Flash | 23.6 |
| Grok 4.7 | 23.2 |
| Opus 4.8 | 22.8 |
| Gemini 3.8 Flash | 22.8 |
| GPT-6 Luna | 22.8 |
| Qwen3.8 Max (0902) | 22.8 |
| GPT-5.5 | 22.6 |
| Kimi K3 | 22.0 |
| Gemini 4 Argon | 21.8 |
| Claude Opus 5 | 21.6 |
| Claude Haiku 5.5 | 20.8 |
| MiniMax M3.1 Flash | 20.8 |
| GPT-5.5 Instant (2026-06-25 hosted snapshot) | 20.2 |
| Qwen3.8-Max | 20.2 |
| Gemini 3.5 Flash | 19.8 |
| MiMo-V2.6-Pro | 19.2 |
| Grok 4.5 | 18.8 |
| Mistral Large 4 Preview | 18.6 |
| Muse Spark Open | 18.6 |
| Gemini 3.1 Pro Preview | 17.8 |
| Grok 4.6 | 17.8 |
| Gemini 3.6 Flash | 17.4 |
| Muse Spark 1.2 | 17.4 |
| DeepSeek V4.1 Flash 1017 | 17.2 |
| Qwen3.8 27B | 16.4 |