General Assistant Tasks on GAIA L2
49.8AccuracyExComm
Evaluation Results
| Method | Links | |
|---|---|---|
| ExCommBackbone=Qwen2.5-7B, N > 1?=true, Available Tools=C / F / S2026.05 | 49.8 | |
| Independent ScalingBackbone=Qwen2.5-7B, N > 1?=true, Available Tools=C / F / S2026.05 | 46.8 | |
| Tree SearchBackbone=Qwen2.5-7B, N > 1?=true, Available Tools=C / F / S2026.05 | 45.3 | |
| Independent Scaling + SRBackbone=Qwen2.5-7B, N > 1?=true, Available Tools=C / F / S2026.05 | 43.7 | |
| Base AgentBackbone=Qwen2.5-7B, N > 1?=false, Available Tools=C / F / S2026.05 | 38.8 | |
| Sequential Revision (SR)Backbone=Qwen2.5-7B, N > 1?=false, Available Tools=C / F / S2026.05 | 36.7 | |
| ExCommBackbone=Gemini-2.5-Flash-Lite, N > 1?=true, Available Tools=C / F / S2026.05 | 30.8 | |
| Tree SearchBackbone=Gemini-2.5-Flash-Lite, N > 1?=true, Available Tools=C / F / S2026.05 | 30 | |
| Independent Scaling + SRBackbone=Gemini-2.5-Flash-Lite, N > 1?=true, Available Tools=C / F / S2026.05 | 29.2 | |
| Independent ScalingBackbone=Gemini-2.5-Flash-Lite, N > 1?=true, Available Tools=C / F / S2026.05 | 28.5 | |
| Sequential Revision (SR)Backbone=Gemini-2.5-Flash-Lite, N > 1?=false, Available Tools=C / F / S2026.05 | 21.8 | |
| Base AgentBackbone=Gemini-2.5-Flash-Lite, N > 1?=false, Available Tools=C / F / S2026.05 | 21.5 |