General AI Assistant Reasoning on GAIA
67.4Pass@1 AccuracyOPENAI DEEPRESEARCH
Evaluation Results
| Method | Links | |
|---|---|---|
| OPENAI DEEPRESEARCHCategory=CLOSED-SOURCE2026.03 | 67.4 | |
| MIRO-30B + WEDASCategory=OPEN-SOURCE2026.03 | 66.99 | |
| MIRO-30B + MIROFLOWCategory=OPEN-SOURCE2026.03 | 63.11 | |
| GPT-5-MINI + WEDASCategory=OPEN-SOURCE2026.03 | 57.28 | |
| WEBSAILOR-72BCategory=OPEN-SOURCE2026.03 | 55.4 | |
| WEBSAILOR-32BCategory=OPEN-SOURCE2026.03 | 53.2 | |
| ASEARCHER-WEB-32BCategory=OPEN-SOURCE2026.03 | 52.8 | |
| WEBDANCER-QWQ-32BCategory=OPEN-SOURCE2026.03 | 51.5 | |
| GPT-5-MINI + MIROFLOWCategory=OPEN-SOURCE2026.03 | 51.46 | |
| WEBTHINKER-32B-RLCategory=OPEN-SOURCE2026.03 | 48.5 | |
| SEARCH-O1-32BCategory=OPEN-SOURCE2026.03 | 39.8 | |
| O4-MINICategory=DIRECT INFERENCE2026.03 | 33.3 | |
| GPT-4.1Category=DIRECT INFERENCE2026.03 | 22.3 | |
| QWQ-32BCategory=DIRECT INFERENCE2026.03 | 22.3 | |
| GPT-4OCategory=DIRECT INFERENCE2026.03 | 17.5 | |
| QWEN-2.5-72BCategory=DIRECT INFERENCE2026.03 | 14.6 | |
| QWEN-2.5-32BCategory=DIRECT INFERENCE2026.03 | 13.6 |