AA-AnalystAgent (pass^5)
Professional Work
Artificial Analysis' private 80-question spreadsheet and document analysis benchmark across 14 domains. The headline pass^5 is the fraction of questions answered correctly on all five independent attempts with the Stirrup agent harness.
Top models (higher is better)
| Model | Score |
|---|---|
| Gemini 3.7 Flash | 60.0 |
| Claude Fable 5.1 | 57.5 |
| Claude Sonnet 5.5 | 57.5 |
| Claude Opus 5.5 | 56.3 |
| Grok 4.7 | 55.0 |
| Claude Haiku 5.5 | 53.8 |
| Claude Opus 5 | 53.8 |
| GPT-6 Astra | 51.3 |
| GPT-5.5 | 50.0 |
| GPT-6.1 Sol | 50.0 |
| Claude Fable 5 | 48.8 |
| Gemini 4 Argon | 48.8 |
| GPT-5.6 Sol | 47.5 |
| Sonnet 5 | 46.3 |
| Gemini 3.8 Flash | 46.3 |
| Opus 4.8 | 45.0 |
| Gemini 3.5 Flash | 45.0 |
| Qwen3.8 Max (0902) | 45.0 |
| Opus 4.7 | 43.8 |
| GPT-6 Luna | 43.8 |
| Gemini 3.1 Pro Preview | 41.3 |
| Grok 4.6 | 41.3 |
| GLM-5.3 | 38.8 |
| Kimi K3.1 Pro | 38.8 |
| Kimi K3 | 38.8 |
| MiMo-V2.6-Pro | 37.5 |
| Muse Spark 1.3 | 37.5 |
| DeepSeek V4.1 Flash | 35.0 |
| Grok 4.5 | 35.0 |
| Step 5 Preview | 35.0 |
| Inkling-Small | 27.5 |
| DeepSeek-V4-Flash | 25.0 |
| GLM-5.3 Flash | 25.0 |
| Inkling | 23.8 |
| Sonnet 4.6 | 20.0 |
| MiMo-V2.5-Pro | 20.0 |
| Muse Glimmer 30B | 20.0 |
| DeepSeek-V4-Pro | 18.8 |
| Qwen3.7-Max (source unspecified) | 18.8 |
| Ling 3.0 Flash Fin | 16.3 |