# Vals MortgageTax (MortgageTax)

> Vals AI benchmark for mortgage and tax document reasoning, including semantic and numerical extraction task views.

Canonical page: https://benchlm.ai/benchmarks/valsmortgagetax

- Category: [Agentic](/agentic)
- Last updated: September 1, 2026

## About MortgageTax

- Year: 2026
- Tasks: Mortgage and tax extraction tasks
- Format: Accuracy score
- Difficulty: Professional mortgage-tax document reasoning
- Paper: [MortgageTax](https://www.vals.ai/benchmarks/mortgage_tax)

BenchLM mirrors Vals MortgageTax as a display-only finance and document-reasoning benchmark.

MortgageTax is currently displayed on BenchLM for reference, but it is excluded from the weighted scoring formula.

## Leaderboard (98 models)

| Rank | Model | Configuration | Creator | Score |
|------|-------|---------------|---------|-------|
| 1 | [Claude Opus 5](/models/claude-opus-5) | — | Anthropic | 72.06% |
| 2 | [Claude Fable 5.1](/models/claude-fable-5-1) | — | Anthropic | 70.79% |
| 3 | [Claude Opus 4.7](/models/claude-opus-4-7) | — | Anthropic | 70.27% |
| 4 | [Claude Sonnet 5](/models/claude-sonnet-5) | — | Anthropic | 70.03% |
| 5 | [Claude Opus 4.8](/models/claude-opus-4-8) | — | Anthropic | 69.91% |
| 6 | [Gemini 3.1 Pro Preview](https://www.vals.ai/models/google_gemini-3.1-pro-preview) | high reasoning | Google | 69.40% |
| 7 | [Gemini 3 Pro Preview](https://www.vals.ai/models/google_gemini-3-pro-preview) | high reasoning | Google | 69.08% |
| 8 | [Claude Fable 5](/models/claude-fable) | — | Anthropic | 68.92% |
| 9 | [Gemini 2.5 Pro](/models/gemini-2-5-pro) | — | Google | 68.92% |
| 10 | [GPT-5.5](/models/gpt-5-5) | xhigh reasoning | OpenAI | 68.76% |
| 11 | [Gemini 3 Flash Preview](https://www.vals.ai/models/google_gemini-3-flash-preview) | high reasoning | Google | 68.72% |
| 12 | [Gemini 3.5 Flash-Lite](/models/gemini-3-5-flash-lite) | high reasoning | Google | 68.68% |
| 13 | [Claude 3.7 Sonnet](https://www.vals.ai/models/anthropic_claude-3-7-sonnet-20250219) | — | Anthropic | 68.68% |
| 14 | [Claude Opus 4.5](/models/claude-opus-4-5) | — | Anthropic | 68.68% |
| 15 | [Claude Opus 4.6 (Adaptive)](/models/claude-opus-4-6-thinking) | — | Anthropic | 68.52% |
| 16 | [MiniMax M3](/models/minimax-m3) | — | MiniMax | 68.36% |
| 17 | [GPT-5.4](/models/gpt-5-4) | xhigh reasoning | OpenAI | 68.32% |
| 18 | [Qwen3.6-27B](/models/qwen3-6-27b) | — | Alibaba | 68.28% |
| 19 | [Gemini 3.5 Flash](/models/gemini-3-5-flash) | high reasoning | Google | 68.12% |
| 20 | [Gemini 3.1 Flash Lite Preview](https://www.vals.ai/models/google_gemini-3.1-flash-lite-preview) | high reasoning | Google | 68.04% |
| 21 | [Qwen3.6 Plus](/models/qwen3-6-plus) | — | Alibaba | 67.97% |
| 22 | [Claude Sonnet 4.6](/models/claude-sonnet-4-6) | — | Anthropic | 67.73% |
| 23 | [Claude Opus 4.5 Thinking](/models/claude-opus-4-5-thinking) | — | Anthropic | 67.69% |
| 24 | [Gemini 3.6 Flash](/models/gemini-3-6-flash) | high reasoning | Google | 67.61% |
| 25 | [Qwen3.5 Flash](/models/qwen3-5-flash) | — | Alibaba | 67.37% |
| 26 | [GPT-5.6 Terra](/models/gpt-5-6-terra) | xhigh reasoning | OpenAI | 67.33% |
| 27 | [GPT-5.6 Luna](/models/gpt-5-6-luna) | max reasoning | OpenAI | 67.29% |
| 28 | [GPT-5.6 Sol](/models/gpt-5-6-sol) | max reasoning | OpenAI | 67.29% |
| 29 | [Gemini 2.5 Pro Exp 03 25](https://www.vals.ai/models/google_gemini-2.5-pro-exp-03-25) | — | Google | 67.17% |
| 30 | [GPT-5.2](/models/gpt-5-2) | xhigh reasoning | OpenAI | 67.13% |
| 31 | [GPT-5 mini](/models/gpt-5-mini) | high reasoning | OpenAI | 66.89% |
| 32 | [Claude 3.7 Sonnet 20250219 Thinking](https://www.vals.ai/models/anthropic_claude-3-7-sonnet-20250219-thinking) | — | Anthropic | 66.85% |
| 33 | [Gemini 3.7 Flash](/models/gemini-3-7-flash) | high reasoning | Google | 66.65% |
| 34 | [Kimi K2.5 Thinking](https://www.vals.ai/models/kimi_kimi-k2.5-thinking) | — | Moonshot AI | 66.53% |
| 35 | [Kimi K3](/models/kimi-k3) | — | Moonshot AI | 66.34% |
| 36 | [Qwen3.7 Plus](/models/qwen3-7-plus) | — | Alibaba | 66.18% |
| 37 | [Muse Spark 1.1](/models/muse-spark-1-1) | xhigh reasoning | Meta | 66.14% |
| 38 | [GPT-4.1](/models/gpt-4-1) | high reasoning | OpenAI | 65.94% |
| 39 | [Kimi K2.6](/models/kimi-2-6) | — | Moonshot AI | 65.82% |
| 40 | [o3](/models/o3) | high reasoning | OpenAI | 65.70% |
| 41 | [GPT-4.1 mini](/models/gpt-4-1-mini) | high reasoning | OpenAI | 65.50% |
| 42 | [GPT-5](https://www.vals.ai/models/openai_gpt-5-2025-08-07) | high reasoning | OpenAI | 65.45% |
| 43 | [Muse Spark 1.2](/models/muse-spark-1-2) | xhigh reasoning | Meta | 65.42% |
| 44 | [Gemini 3.8 Flash](/models/gemini-3-8-flash) | high reasoning | Google | 65.34% |
| 45 | [Qwen3.8-27B](/models/qwen3-8-27b) | xhigh reasoning | Alibaba | 64.94% |
| 46 | [O4 Mini](https://www.vals.ai/models/openai_o4-mini-2025-04-16) | high reasoning | OpenAI | 64.83% |
| 47 | [Grok 4.6](/models/grok-4-6) | high reasoning | xAI | 64.19% |
| 48 | [Claude 3.5 Sonnet](/models/claude-3-5-sonnet) | — | Anthropic | 64.07% |
| 49 | [Qwen3.8 Max](/models/qwen3-8-max) | — | Alibaba | 63.99% |
| 50 | [Claude Sonnet 4.5 Thinking](/models/claude-sonnet-4-5-thinking) | — | Anthropic | 63.99% |
| 51 | [Inkling](/models/inkling) | 0.99 reasoning | Thinking Machines Lab | 63.99% |
| 52 | [GPT-5.4 mini](/models/gpt-5-4-mini) | xhigh reasoning | OpenAI | 63.51% |
| 53 | [Gemini 2.5 Flash Preview 09 2025](https://www.vals.ai/models/google_gemini-2.5-flash-preview-09-2025) | — | Google | 62.60% |
| 54 | [Claude Sonnet 4](https://www.vals.ai/models/anthropic_claude-sonnet-4-20250514) | — | Anthropic | 62.47% |
| 55 | [Inkling-Small](/models/inkling-small) | 0.99 reasoning | Thinking Machines Lab | 62.28% |
| 56 | [Claude Haiku 4.5 Thinking](/models/claude-haiku-4-5-thinking) | — | Anthropic | 62.16% |
| 57 | [Magistral Small 2509](https://www.vals.ai/models/mistralai_magistral-small-2509) | — | Mistral AI | 62.12% |
| 58 | [Gemini 2.5 Flash Preview 09 2025 Thinking](https://www.vals.ai/models/google_gemini-2.5-flash-preview-09-2025-thinking) | — | Google | 61.92% |
| 59 | [Grok 4.5](/models/grok-4-5) | high reasoning | xAI | 61.80% |
| 60 | [Gemma 4 31b It](https://www.vals.ai/models/google_gemma-4-31b-it) | high reasoning | Google | 61.37% |
| 61 | [GPT-5.1](/models/gpt-5-1) | high reasoning | OpenAI | 61.37% |
| 62 | [Mistral Small 2503](https://www.vals.ai/models/mistralai_mistral-small-2503) | — | Mistral AI | 61.21% |
| 63 | [Claude Opus 4.1](https://www.vals.ai/models/anthropic_claude-opus-4-1-20250805) | — | Anthropic | 61.09% |
| 64 | [GPT-4o](/models/gpt-4o) | high reasoning | OpenAI | 60.97% |
| 65 | [Qwen3.5 Plus Thinking](https://www.vals.ai/models/alibaba_qwen3.5-plus-thinking) | — | Alibaba | 60.77% |
| 66 | [Gemini 2.0 Flash Exp](https://www.vals.ai/models/google_gemini-2.0-flash-exp) | — | Google | 60.37% |
| 67 | [Gemini 2.0 Flash 001](https://www.vals.ai/models/google_gemini-2.0-flash-001) | — | Google | 59.66% |
| 68 | [MiMo-V2.5](/models/mimo-v2-5) | — | Xiaomi | 59.26% |
| 69 | [GPT-5.4 nano](/models/gpt-5-4-nano) | high reasoning | OpenAI | 59.10% |
| 70 | [Claude Opus 4](https://www.vals.ai/models/anthropic_claude-opus-4-20250514) | — | Anthropic | 58.59% |
| 71 | [Llama4 Maverick Instruct Basic](https://www.vals.ai/models/fireworks_llama4-maverick-instruct-basic) | — | Fireworks AI | 58.51% |
| 72 | [Gemini 2.5 Flash Preview 04 17](https://www.vals.ai/models/google_gemini-2.5-flash-preview-04-17) | — | Google | 58.43% |
| 73 | [Meta Llama Llama 4 Scout 17B 16E Instruct](https://www.vals.ai/models/together_meta-llama/Llama-4-Scout-17B-16E-Instruct) | — | Together AI | 57.75% |
| 74 | [Gemini 2.5 Flash Lite Preview 09 2025 Thinking](https://www.vals.ai/models/google_gemini-2.5-flash-lite-preview-09-2025-thinking) | — | Google | 57.55% |
| 75 | [Gemini 2.5 Flash Preview 04 17 Thinking](https://www.vals.ai/models/google_gemini-2.5-flash-preview-04-17-thinking) | — | Google | 57.51% |
| 76 | [GPT-4o](/models/gpt-4o) | high reasoning | OpenAI | 57.43% |
| 77 | [Gemini 1.5 Pro 002](https://www.vals.ai/models/google_gemini-1.5-pro-002) | — | Google | 56.76% |
| 78 | [Claude Opus 4.1 20250805 Thinking](https://www.vals.ai/models/anthropic_claude-opus-4-1-20250805-thinking) | — | Anthropic | 56.12% |
| 79 | [Gemini 2.5 Flash Lite Preview 09 2025](https://www.vals.ai/models/google_gemini-2.5-flash-lite-preview-09-2025) | — | Google | 55.80% |
| 80 | [GPT-4o mini](/models/gpt-4o-mini) | high reasoning | OpenAI | 54.49% |
| 81 | [GPT-5 nano](/models/gpt-5-nano) | high reasoning | OpenAI | 53.62% |
| 82 | [GPT-4.1 nano](/models/gpt-4-1-nano) | high reasoning | OpenAI | 52.82% |
| 83 | [Mistral Large 2512](https://www.vals.ai/models/mistralai_mistral-large-2512) | — | Mistral AI | 52.11% |
| 84 | [Magistral Medium 2509](https://www.vals.ai/models/mistralai_magistral-medium-2509) | — | Mistral AI | 51.87% |
| 85 | [Claude Sonnet 4 20250514 Thinking](https://www.vals.ai/models/anthropic_claude-sonnet-4-20250514-thinking) | — | Anthropic | 49.88% |
| 86 | [Grok 4.3](/models/grok-4-3) | — | xAI | 48.25% |
| 87 | [Grok 4.20 0309 Reasoning](https://www.vals.ai/models/grok_grok-4.20-0309-reasoning) | — | xAI | 45.35% |
| 88 | [Grok 4 0709](https://www.vals.ai/models/grok_grok-4-0709) | — | xAI | 44.48% |
| 89 | [Gemini 1.5 Flash 002](https://www.vals.ai/models/google_gemini-1.5-flash-002) | — | Google | 42.77% |
| 90 | [Grok 4.1 Fast (Reasoning)](/models/grok-4-1-fast-reasoning) | — | xAI | 42.61% |
| 91 | [Grok 4 Fast (Reasoning)](/models/grok-4-fast-reasoning) | — | xAI | 42.09% |
| 92 | [Meta Llama Llama 3.2 90B Vision Instruct Turbo](https://www.vals.ai/models/together_meta-llama/Llama-3.2-90B-Vision-Instruct-Turbo) | — | Together AI | 38.79% |
| 93 | [Mistral Medium 2505](https://www.vals.ai/models/mistralai_mistral-medium-2505) | — | Mistral AI | 36.45% |
| 94 | [Grok 4.1 Fast Non Reasoning](https://www.vals.ai/models/grok_grok-4-1-fast-non-reasoning) | — | xAI | 34.26% |
| 95 | [Mistral Medium 3.5](https://www.vals.ai/models/mistralai_mistral-medium-3.5) | high reasoning | Mistral AI | 28.89% |
| 96 | [Grok 4 Fast Non Reasoning](https://www.vals.ai/models/grok_grok-4-fast-non-reasoning) | — | xAI | 25.32% |
| 97 | [Meta Llama Llama 3.2 11B Vision Instruct Turbo](https://www.vals.ai/models/together_meta-llama/Llama-3.2-11B-Vision-Instruct-Turbo) | — | Together AI | 23.29% |
| 98 | [Grok 2 Vision 1212](https://www.vals.ai/models/grok_grok-2-vision-1212) | — | xAI | 3.46% |

## FAQ

### What does MortgageTax measure?

Vals AI benchmark for mortgage and tax document reasoning, including semantic and numerical extraction task views.

### Which model leads the published MortgageTax snapshot?

Claude Opus 5 currently leads the published MortgageTax snapshot with a score of 72.06%.

### How many models are evaluated on MortgageTax?

The September 1, 2026 contains 98 AI models.

### Does MortgageTax affect BenchLM's overall score?

Not directly. MortgageTax is still displayed on BenchLM for reference, but it is excluded from the weighted scoring formula.