Skip to content

Benchmark

SpreadsheetBench 2 Debug

Agentic benchmarks measure how well models carry out tasks that take several steps. Scores are percentages. Higher is better.

Models measured
7
Scores recorded
7
Sources
1
IQ weight
0.34

Score distribution

6.9%
Mean
7%
Median
3.1%
Std dev
4%
10th percentile
10%
90th percentile
3%
Min
3%12%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 0.34 for chat models, measured over 7 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
1Claude Opus 4.6scaffold: swe-agent, date: 2026-03-26· SpreadsheetBench 2 leaderboard data12%
2DeepSeek V3.2scaffold: swe-agent, date: 2026-03-26· SpreadsheetBench 2 leaderboard data10%
3GPT 5.2scaffold: swe-agent, date: 2026-03-26· SpreadsheetBench 2 leaderboard data8%
4Gemini 3.1 Pro Previewscaffold: swe-agent, date: 2026-03-26· SpreadsheetBench 2 leaderboard data7%
5Kimi K2.5scaffold: swe-agent, date: 2026-03-26· SpreadsheetBench 2 leaderboard data4%
6MiniMax M2.5scaffold: swe-agent, date: 2026-03-26· SpreadsheetBench 2 leaderboard data4%
7Qwen3.5 397B A17Bscaffold: swe-agent, date: 2026-03-26· SpreadsheetBench 2 leaderboard data3%

Data credits: SpreadsheetBench (Renmin University of China and contributors), spreadsheetbench.github.io.