Skip to main content

Vals MortgageTax (MortgageTax)

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

How BenchLM shows MortgageTax

BenchLM mirrors the public Vals AI MortgageTax leaderboard captured from https://www.vals.ai/benchmarks/mortgage_tax and updated by Vals on May 16, 2026. The snapshot preserves overall scores, uncertainty, latency, cost-per-test metadata, and task-level scores where Vals publishes them.

MortgageTax is display only on BenchLM. Vals proprietary or Vals-hosted aggregate views are useful context, but BenchLM does not use them as weighted ranking inputs or as a replacement for benchmark-native source records.

74 Vals rows3 task viewsprivate datasetTasks: Overall, Semantic Extraction, Numerical ExtractionDisplay only

MortgageTax score on MortgageTax — May 16, 2026

BenchLM mirrors the published mortgagetax score view for MortgageTax. Claude Opus 4.7 leads the public snapshot at 70.27% , followed by Gemini 3.1 Pro Preview (69.40%) and Gemini 3 Pro Preview (69.08%). BenchLM does not use these results to rank models overall.

74 modelsExternal benchmark mirrorsCurrentDisplay onlyUpdated May 16, 2026

The published MortgageTax snapshot is tightly clustered at the top: Claude Opus 4.7 sits at 70.27%, while the third row is only 1.19 points behind. The broader top-10 spread is 1.95 points, so many of the published scores sit in a relatively narrow band.

74 models have been evaluated on MortgageTax. The benchmark falls in the External benchmark mirrors category. BenchLM tracks this category separately from its weighted global scoring system, so these results are best compared on the dedicated Korean benchmark views. MortgageTax is currently displayed for reference but excluded from the scoring formula, so it does not directly affect overall rankings.

About MortgageTax

Year

2026

Tasks

Mortgage and tax extraction tasks

Format

Accuracy score

Difficulty

Professional mortgage-tax document reasoning

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

BenchLM freshness & provenance

Version

MortgageTax 2026

Refresh cadence

Quarterly

Staleness state

Current

Question availability

Public benchmark set

CurrentDisplay only

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.

MortgageTax score table (74 models)

1
Claude Opus 4.7anthropic/claude-opus-4-7
70.27%
2
Gemini 3.1 Pro Previewgoogle/gemini-3.1-pro-preview
69.40%
3
Gemini 3 Pro Previewgoogle/gemini-3-pro-preview
69.08%
4
Gemini 2.5 Progoogle/gemini-2.5-pro
68.92%
5
GPT-5.5openai/gpt-5.5
68.76%
6
Gemini 3 Flash Previewgoogle/gemini-3-flash-preview
68.72%
7
Claude 3.7 Sonnetanthropic/claude-3-7-sonnet-20250219
68.68%
8
Claude Opus 4.5anthropic/claude-opus-4-5-20251101
68.68%
9
Claude Opus 4.6 Thinkinganthropic/claude-opus-4-6-thinking
68.52%
10
GPT-5.4openai/gpt-5.4-2026-03-05
68.32%
11
Qwen3.6 27balibaba/qwen3.6-27b
68.28%
12
Gemini 3.5 Flashgoogle/gemini-3.5-flash
68.12%
13
Gemini 3.1 Flash Lite Previewgoogle/gemini-3.1-flash-lite-preview
68.04%
14
Qwen3.6 Plusalibaba/qwen3.6-plus
67.97%
15
Claude Sonnet 4.6anthropic/claude-sonnet-4-6
67.73%
16
Claude Opus 4.5 20251101 Thinkinganthropic/claude-opus-4-5-20251101-thinking
67.69%
17
Qwen3.5 Flashalibaba/qwen3.5-flash
67.37%
18
Gemini 2.5 Pro Exp 03 25google/gemini-2.5-pro-exp-03-25
67.17%
19
GPT-5.2openai/gpt-5.2-2025-12-11
67.13%
20
GPT-5 Miniopenai/gpt-5-mini-2025-08-07
66.89%
21
Claude 3.7 Sonnet 20250219 Thinkinganthropic/claude-3-7-sonnet-20250219-thinking
66.85%
22
Kimi K2.5 Thinkingkimi/kimi-k2.5-thinking
66.53%
23
GPT-4.1openai/gpt-4.1-2025-04-14
65.94%
24
Kimi K2.6 Thinkingkimi/kimi-k2.6-thinking
65.82%
25
O3openai/o3-2025-04-16
65.70%
26
GPT-4.1 Miniopenai/gpt-4.1-mini-2025-04-14
65.50%
27
GPT-5openai/gpt-5-2025-08-07
65.45%
28
O4 Miniopenai/o4-mini-2025-04-16
64.83%
29
Claude 3.5 Sonnetanthropic/claude-3-5-sonnet-20241022
64.07%
30
Claude Sonnet 4.5 20250929 Thinkinganthropic/claude-sonnet-4-5-20250929-thinking
63.99%
31
GPT-5.4 Miniopenai/gpt-5.4-mini-2026-03-17
63.51%
32
Gemini 2.5 Flash Preview 09 2025google/gemini-2.5-flash-preview-09-2025
62.60%
33
Claude Sonnet 4anthropic/claude-sonnet-4-20250514
62.47%
34
Claude Haiku 4.5 20251001 Thinkinganthropic/claude-haiku-4-5-20251001-thinking
62.16%
35
Magistral Small 2509mistralai/magistral-small-2509
62.12%
36
Gemini 2.5 Flash Preview 09 2025 Thinkinggoogle/gemini-2.5-flash-preview-09-2025-thinking
61.92%
37
Gemma 4 31b Itgoogle/gemma-4-31b-it
61.37%
38
GPT-5.1openai/gpt-5.1-2025-11-13
61.37%
39
Mistral Small 2503mistralai/mistral-small-2503
61.21%
40
Claude Opus 4.1anthropic/claude-opus-4-1-20250805
61.09%
41
GPT-4oopenai/gpt-4o-2024-08-06
60.97%
42
Qwen3.5 Plus Thinkingalibaba/qwen3.5-plus-thinking
60.77%
43
Gemini 2.0 Flash Expgoogle/gemini-2.0-flash-exp
60.37%
44
Gemini 2.0 Flash 001google/gemini-2.0-flash-001
59.66%
45
GPT-5.4 Nanoopenai/gpt-5.4-nano-2026-03-17
59.10%
46
Claude Opus 4anthropic/claude-opus-4-20250514
58.59%
47
Llama4 Maverick Instruct Basicfireworks/llama4-maverick-instruct-basic
58.51%
48
Gemini 2.5 Flash Preview 04 17google/gemini-2.5-flash-preview-04-17
58.43%
49
Meta Llama Llama 4 Scout 17B 16E Instructtogether/meta-llama/Llama-4-Scout-17B-16E-Instruct
57.75%
50
Gemini 2.5 Flash Lite Preview 09 2025 Thinkinggoogle/gemini-2.5-flash-lite-preview-09-2025-thinking
57.55%
51
Gemini 2.5 Flash Preview 04 17 Thinkinggoogle/gemini-2.5-flash-preview-04-17-thinking
57.51%
52
GPT-4oopenai/gpt-4o-2024-11-20
57.43%
53
Gemini 1.5 Pro 002google/gemini-1.5-pro-002
56.76%
54
Claude Opus 4.1 20250805 Thinkinganthropic/claude-opus-4-1-20250805-thinking
56.12%
55
Gemini 2.5 Flash Lite Preview 09 2025google/gemini-2.5-flash-lite-preview-09-2025
55.80%
56
GPT-4o Miniopenai/gpt-4o-mini-2024-07-18
54.49%
57
GPT-5 Nanoopenai/gpt-5-nano-2025-08-07
53.62%
58
GPT-4.1 Nanoopenai/gpt-4.1-nano-2025-04-14
52.82%
59
Mistral Large 2512mistralai/mistral-large-2512
52.11%
60
Magistral Medium 2509mistralai/magistral-medium-2509
51.87%
61
Claude Sonnet 4 20250514 Thinkinganthropic/claude-sonnet-4-20250514-thinking
49.88%
62
Grok 4.3grok/grok-4.3
48.25%
63
Grok 4.20 0309 Reasoninggrok/grok-4.20-0309-reasoning
45.35%
64
Grok 4 0709grok/grok-4-0709
44.48%
65
Gemini 1.5 Flash 002google/gemini-1.5-flash-002
42.77%
66
Grok 4.1 Fast Reasoninggrok/grok-4-1-fast-reasoning
42.61%
67
Grok 4 Fast Reasoninggrok/grok-4-fast-reasoning
42.09%
68
Meta Llama Llama 3.2 90B Vision Instruct Turbotogether/meta-llama/Llama-3.2-90B-Vision-Instruct-Turbo
38.79%
69
Mistral Medium 2505mistralai/mistral-medium-2505
36.45%
70
Grok 4.1 Fast Non Reasoninggrok/grok-4-1-fast-non-reasoning
34.26%
71
Mistral Medium 3.5mistralai/mistral-medium-3.5
28.89%
72
Grok 4 Fast Non Reasoninggrok/grok-4-fast-non-reasoning
25.32%
73
Meta Llama Llama 3.2 11B Vision Instruct Turbotogether/meta-llama/Llama-3.2-11B-Vision-Instruct-Turbo
23.29%
74
Grok 2 Vision 1212grok/grok-2-vision-1212
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 4.7 currently leads the published MortgageTax snapshot with 70.27% mortgagetax score. BenchLM shows this benchmark for display only and does not use it in overall rankings.

How many models are evaluated on MortgageTax?

74 AI models are included in BenchLM's mirrored MortgageTax snapshot, based on the public leaderboard captured on May 16, 2026.

Last updated: May 16, 2026 · mirrored from the public benchmark leaderboard

The AI models change fast. We track them for you.

For engineers, researchers, and the plain curious — a weekly brief on new models, ranking shifts, and pricing changes.

Free. No spam. Unsubscribe anytime.