Which AI model scores highest on TaxEval v2?

Muse Spark 1.1 by Meta holds the best TaxEval v2 result among tracked models, at 79.72% (released Jul 9 2026). Higher scores are better on this benchmark.

What are the top 4 models on TaxEval v2?

1. Muse Spark 1.1 (Meta) — 79.72%; 2. Claude Fable 5 (Anthropic) — 76.94%; 3. Claude Opus 4.8 (Anthropic) — 75.63%; 4. GPT-5.5 (OpenAI) — 74.98%.

How many models have a published TaxEval v2 score?

4 tracked models have a published TaxEval v2 score. Scores are collected from published lab reports and benchmark sources, with recorded sources credited alongside the results. We do not re-run the evaluations.

Tax questions

TaxEval v2

A set of real tax questions created by Vals AI — can the AI give accurate answers about tax rules and filings? Higher is better.

Rankings

Higher is better

TaxEval v2 — frequently asked questions

: A set of real tax questions created by Vals AI — can the AI give accurate answers about tax rules and filings? Higher is better.
: Muse Spark 1.1 by Meta holds the best TaxEval v2 result among tracked models, at 79.72% (released Jul 9 2026). Higher scores are better on this benchmark.
: 1. Muse Spark 1.1 (Meta) — 79.72%; 2. Claude Fable 5 (Anthropic) — 76.94%; 3. Claude Opus 4.8 (Anthropic) — 75.63%; 4. GPT-5.5 (OpenAI) — 74.98%.
: 4 tracked models have a published TaxEval v2 score. Scores are collected from published lab reports and benchmark sources, with recorded sources credited alongside the results. We do not re-run the evaluations.