Spreadsheet Tasks on SSB (SpreadsheetBench)
100Win RateIG
Evaluation Results
| Method | Links | |
|---|---|---|
| IGOpponent=RLM2026.05 | 100 | |
| IGOpponent=Single-Agent Coding2026.05 | 100 | |
| CC SubagentsOpponent=Single-Agent Coding2026.05 | 100 | |
| CC SubagentsOpponent=RLM2026.05 | 94 | |
| Trace2SkillOpponent=RLM2026.05 | 89 | |
| Trace2SkillOpponent=Single-Agent Coding2026.05 | 89 | |
| IGOpponent=CC Subagents2026.05 | 56 | |
| RLMOpponent=Single-Agent Coding2026.05 | 56 | |
| IGOpponent=Trace2Skill2026.05 | 50 | |
| CC SubagentsOpponent=Trace2Skill2026.05 | 50 | |
| Trace2SkillOpponent=IG2026.05 | 50 | |
| Trace2SkillOpponent=CC Subagents2026.05 | 50 | |
| CC SubagentsOpponent=IG2026.05 | 44 | |
| Single-Agent CodingOpponent=RLM2026.05 | 44 | |
| RLMOpponent=Trace2Skill2026.05 | 11 | |
| Single-Agent CodingOpponent=Trace2Skill2026.05 | 11 | |
| RLMOpponent=CC Subagents2026.05 | 6 | |
| RLMOpponent=IG2026.05 | 0 | |
| Single-Agent CodingOpponent=IG2026.05 | 0 | |
| Single-Agent CodingOpponent=CC Subagents2026.05 | 0 |