General AI Assistant Task Execution on GAIA Round 1
25Level 1 Evo. Pass@1KIMI-K2
Evaluation Results
| Method | Links | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| KIMI-K22025.10 | 25 | 37.7 | -0.127 | 18.6 | 23.3 | -0.047 | 10 | 7.7 | 0.023 | 19.2 | 25.5 | -0.063 | |
| GPT-5-Mini2025.10 | 25 | 56.6 | -0.316 | 27.9 | 46.5 | -0.186 | 20 | 19.2 | 0.008 | 26 | 45.5 | -0.213 | |
| Deepseek-v3.12025.10 | 20 | 50.9 | -0.309 | 27.9 | 41.9 | -0.14 | 20 | 23.1 | -0.031 | 24.7 | 41.8 | -0.171 | |
| Gemini-2.5-flash2025.10 | 20 | 47.1 | -0.271 | 16.3 | 23.2 | -0.069 | 0 | 11.5 | -0.115 | 15.1 | 29.1 | -0.14 | |
| Deepseek-v3.12025.10 | 20 | 50.9 | -0.309 | 25.3 | 41.9 | -0.166 | 11.7 | 23.1 | -0.231 | 21.8 | 41.8 | -0.2 | |
| GPT-5-Mini2025.10 | 20 | 56.6 | -0.366 | 21.5 | 46.5 | -0.25 | 27.8 | 19.2 | 0.086 | 21.8 | 45.5 | -0.237 | |
| KIMI-K22025.10 | 15.6 | 37.7 | -0.221 | 21.5 | 23.3 | -0.018 | 11.1 | 7.7 | 0.034 | 18.3 | 25.5 | -0.072 | |
| Gemini-2.5-flash2025.10 | 11.1 | 47.1 | -0.36 | 17.7 | 23.2 | -0.055 | 5.6 | 11.5 | -0.059 | 14.1 | 29.1 | -0.15 |