Benchmark
SpreadSheetBench-v1
Agentic benchmarks measure how well models carry out tasks that take several steps. Scores are percentages. Higher is better.
- Models measured
- 5
- Scores recorded
- 8
- Sources
- 2
- IQ weight
- 0.00
Score distribution
- 84.5%
- Mean
- 87%
- Median
- 7.7%
- Std dev
- 86.3%
- 10th percentile
- 89.4%
- 90th percentile
- 64.3%
- Min
This benchmark has a weight of 0.00 for chat models, measured over 5 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 | MiniMax M3· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 89.4% |
| 2 | MiniMax M3public dataset · Claude Code· Reported by MiniMax Research· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 89.4% |
| 3 | Qwen3.7 Max· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 87% |
| 4 | Qwen3.7 MaxQwen3.7-Max release · published setting· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 87% |
| 5 | Ling 3.0 Flash Fin· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 86.5% |
| 6 | Qwen3.7 Plus· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 86.3% |
| 7 | Qwen3.7 PlusQwen3.7-Plus release · published setting· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 86.3% |
| 8 | Claude Opus 4.5full 912 tasks · custom bash/Python/LibreOffice harness · no extended thinking · 200K context · avg@5· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 64.3% |
Data credits: 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. · Data by LLM Stats