Leaderboard wordt geladen
InsureBench tests AI models on a private Wft-Basis practice-question set. Phase 2 expands this to open insurance-advice cases.
Mistral: Mistral Nemo ranks #1 with 21/40 on the combined Wft and prompt score.
The score measures Wft-Basis knowledge, not whether a model is suitable as a standalone AI adviser.
Model X ranks highest on the InsureBench Wft-Basis knowledge benchmark.
Model X gives the best insurance advice in private simple-risk cases.
Whether AI models are reliable enough for standalone insurance advice requires phase 2-3.
Score on a 40-point scale (Wft-Basis equivalent). Click a model for details.
A model may appear in multiple rows — one per benchmark type. WFT measures knowledge (multiple choice), Prompt measures advice skills, Combined measures both.
Scores in the same group differ by less than 1 point and should be read as effectively neck-and-neck.
Mistralai
OSAnthropic
DeepSeek
OSOpenAI
DeepSeek
OSDeepSeek
OSxAI
Moonshotai
OpenAI
OpenAI
| # | Model | Provider | Open source | Score (40) | WFT | Prompt | Price / M tokens | Result | Last tested |
|---|---|---|---|---|---|---|---|---|---|
| 1 | Mistral: Mistral NemoCombined | Mistralai | Open source | 21 / 40 43/80 raw Groep A | 22 / 40 | 19 / 40 | €0.02 in / €0.03 out | Fail -6.2 | 01 May 2026 |
| 2 | Claude Haiku 4.5 |
Each public round now follows the same editorial structure: what this round says, what changed, which outliers are explainable, and what still must not be concluded.
This round makes the Wft-Basis leaderboard citation-grade readable: public field definitions, score groups, and fixed source pages now ship as one release.
Download aggregated run data as CSV or JSON. Use the BibTeX entry below for attribution.
@online{insurebench_wft_basis_1_1_0,
title = {InsureBench: Wft-Basis AI Benchmark},
author = {InsureBench},
year = {2026},
version = {1.1.0},
url = {https://www.insurebench.nl/nl/wft-basis},
urldate = {2026-04-23},
note = {Public leaderboard, 80 questions, 3 runs per model}
}| Anthropic |
| — |
23 / 40 55/80 raw Groep A |
| 28 / 40 |
| 17 / 40 |
| €0.92 in / €4.60 out |
| Fail -4.2 |
| 28 Apr 2026 |
| 3 | DeepSeek-R1Combined | DeepSeek | Open source | 26 / 40 50/80 raw Groep A | 25 / 40 | 26 / 40 | €0.64 in / €2.30 out | Fail -1.2 | 29 Apr 2026 |
| 4 | OpenAI: GPT-5.5 ProCombined | OpenAI | — | 26 / 40 55/80 raw Groep A | 27 / 40 | 25 / 40 | €27.60 in / €165.60 out | Fail -1.2 | 02 May 2026 |
| 5 | DeepSeek V3.2Combined | DeepSeek | Open source | 28 / 40 56/80 raw Groep A | 28 / 40 | 27 / 40 | €0.23 in / €0.35 out | Pass +0.8 | 29 Apr 2026 |
| 6 | DeepSeek: DeepSeek V4 ProCombined | DeepSeek | Open source | 28 / 40 58/80 raw Groep A | 29 / 40 | 26 / 40 | €0.40 in / €0.80 out | Pass +0.8 | 04 May 2026 |
| 7 | Grok 4.20 Beta ReasoningWft | xAI | — | 28 / 40 57/80 raw Groep A | 28 / 40 | — | — | Pass +0.8 | 23 Apr 2026 |
| 8 | MoonshotAI: Kimi K2.5Combined | Moonshotai | — | 28 / 40 60/80 raw Groep A | 30 / 40 | 26 / 40 | €0.40 in / €1.84 out | Pass +0.8 | 06 May 2026 |
| 9 | GPT-4.1Combined | OpenAI | — | 30 / 40 63/80 raw Groep A | 32 / 40 | 28 / 40 | €1.84 in / €7.36 out | Pass +2.8 | 28 Apr 2026 |
| 10 | GPT-4oCombined | OpenAI | — | 30 / 40 57/80 raw Groep A | 29 / 40 | 31 / 40 | €2.30 in / €9.20 out | Pass +2.8 | 29 Apr 2026 |