Benchmark
SpreadsheetBench 2
Agentic benchmarks measure how well models carry out tasks that take several steps. Scores are percentages. Higher is better.
- Models measured
- 11
- Scores recorded
- 12
- Sources
- 3
- IQ weight
- 0.06
Score distribution
- 24.7%
- Mean
- 26.8%
- Median
- 10.1%
- Std dev
- 11.2%
- 10th percentile
- 34.9%
- 90th percentile
- 7.2%
- Min
This benchmark has a weight of 0.06 for chat models, measured over 11 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.
Leaderboard
| # | Model | Score |
|---|---|---|
| 1 | ERNIE 5.5 Flash Preview· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 36.4% |
| 2 | LongCat 2.5· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 34.9% |
| 3 | Claude Opus 4.6scaffold: swe-agent, date: 2026-03-26· SpreadsheetBench 2 leaderboard data | 34.9% |
| 4 | Kimi K3· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 34.8% |
| 5 | Kimi K3max effort · temp 1.0 · top_p 1.0· Reported by Moonshot AI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 34.8% |
| 6 | GPT 5.2scaffold: swe-agent, date: 2026-03-26· SpreadsheetBench 2 leaderboard data | 26.8% |
| 7 | Gemini 3.1 Pro Previewscaffold: swe-agent, date: 2026-03-26· SpreadsheetBench 2 leaderboard data | 23.7% |
| 8 | Ling 3.0 Flash Fin· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 21.8% |
| 9 | DeepSeek V3.2scaffold: swe-agent, date: 2026-03-26· SpreadsheetBench 2 leaderboard data | 15.6% |
| 10 | Kimi K2.5scaffold: swe-agent, date: 2026-03-26· SpreadsheetBench 2 leaderboard data | 14.6% |
| 11 | Qwen3.5 397B A17Bscaffold: swe-agent, date: 2026-03-26· SpreadsheetBench 2 leaderboard data | 11.2% |
| 12 | MiniMax M2.5scaffold: swe-agent, date: 2026-03-26· SpreadsheetBench 2 leaderboard data | 7.2% |
Data credits: SpreadsheetBench (Renmin University of China and contributors), spreadsheetbench.github.io. · Data by LLM Stats · BenchBoard by SYuan03 (github.com/SYuan03/benchboard). MIT covers its code and data organization; no results licence is inferred. Numeric relays retain the original stating-source links.