Finch (FinWorkBench): Finance & Accounting across Spreadsheet-Centric Enterprise Workflows (Finch)
We show this table for reference; we do not rank on it.
A benchmark of real finance and accounting workflows built around spreadsheets: creating, editing and analysing workbooks the way an analyst or accountant would.
Benchmark score on Finch — October 29, 2026
We compile the Finch rows from provider self-reports and secondary reports. Kimi K3 leads the table at 77.3%, followed by Mistral Large 4 (73.3%) and DeepSeek V4 Pro 0813 (67.4%). We do not use these results to rank models overall.
Kimi K3
Moonshot AI
Mistral Large 4
Mistral
DeepSeek V4 Pro 0813
DeepSeek
4 modelsAgenticCurrentDisplay onlyUpdated October 29, 2026
Benchmark score table (4 models)
| Rank | Model / configuration | Score | Parameters (B) | Open / closed |
|---|---|---|---|---|
| 1 | Kimi K3Moonshot AI | 77.3% | Not reported | Open |
| 2 | Mistral Large 4Mistral | 73.3% | Not reported | Open |
| 3 | DeepSeek V4 Pro 0813DeepSeek | 67.4% | Not reported | Open |
| 4 | GLM-5.3Z.AI | 65.1% | Not reported | Open |
Among the reported Finch rows, Kimi K3 is first at 77.3%. The third row is 9.9 points behind. The broader top-10 range is 12.2 points, so the table still separates the published systems.
4 models have been evaluated on Finch. The benchmark falls in the Agentic category. Finch is currently displayed for reference but excluded from the scoring formula, so it does not directly affect overall rankings.
About Finch
Year
2026
Tasks
Spreadsheet-centric finance and accounting workflows
Format
Task score
Difficulty
Enterprise finance work
Mistral reported these results in its October 29, 2026 Mistral Large 4 open-weight post, including the rival-model values in its chart. They are provider-run, so BenchLM shows them display-only and keeps them out of weighted scores.
Freshness and provenance
Version
Finch 2026
Refresh cadence
Quarterly
Staleness state
Current
Question availability
Public benchmark set
BenchLM uses freshness metadata to decide whether a benchmark should still be treated as a strong differentiator, a benchmark to watch, or a display-only reference. For the full scoring policy, see the BenchLM methodology page.
Questions
What does Finch measure?
A benchmark of real finance and accounting workflows built around spreadsheets: creating, editing and analysing workbooks the way an analyst or accountant would.
Which model scores highest on Finch?
Kimi K3 by Moonshot AI currently leads with a score of 77.3% on Finch.
How many models are evaluated on Finch?
4 AI models have published results on Finch in the BenchLM catalog.
Compare top models on Finch
Know when it’s worth switching models
The model to choose, the cheaper alternative, and the release we would wait on.
Read a sample issueJoin 6,100+ subscribers.
One email each week. Unsubscribe anytime.