2026-09-10MODELDeepSeek-V4.1-Flash released with extreme agent memory efficiency and evaluated across VALS • 2026-09-03MODELGPT-6 Astra released by OpenAI - 71.7% on EMB, 63.3% on Tax Agent Bench • 2026-09-02MODELGoogle Gemini 3.8 Flash takes #1 rank on Finance Agent (v2) with 61.4% and 72.2% on EMB • 2026-09-02MODELMeta releases Muse Spark 1.3 & 1.3 Max - 59.0%-60.0% on Finance Agent (v2) • 2026-09-01MODELClaude Fable 5.1 leads Tax Agent Bench (77.6%) and EMB (76.7%) with 75% cheaper cache reads • 2026-09-11BENCHTax Agent Bench and FAB v2 benchmarks updated with latest frontier model rankings • 2026-09-15UPDATEPlatform updated: 43 models tracked across 41 independent benchmarks • 2026-09-10MODELDeepSeek-V4.1-Flash released with extreme agent memory efficiency and evaluated across VALS • 2026-09-03MODELGPT-6 Astra released by OpenAI - 71.7% on EMB, 63.3% on Tax Agent Bench • 2026-09-02MODELGoogle Gemini 3.8 Flash takes #1 rank on Finance Agent (v2) with 61.4% and 72.2% on EMB • 2026-09-02MODELMeta releases Muse Spark 1.3 & 1.3 Max - 59.0%-60.0% on Finance Agent (v2) • 2026-09-01MODELClaude Fable 5.1 leads Tax Agent Bench (77.6%) and EMB (76.7%) with 75% cheaper cache reads • 2026-09-11BENCHTax Agent Bench and FAB v2 benchmarks updated with latest frontier model rankings • 2026-09-15UPDATEPlatform updated: 43 models tracked across 41 independent benchmarks •
Home/Benchmarks/MortgageTax Benchmark
Tier 2numericalforensic

MortgageTax Benchmark

Evaluates financial calculations and regulatory tax computations on complex mortgage agreements.

Variable tasks
17 models evaluated
Updated 2026
VALS (2026)
View Source →Submit your scoreMethodology
72.1%
Top Score
Top model
Claude Opus 5
Model Rankings
financebenchmark.ai · MortgageTax
Anthropic
Anthropic
OpenAI
xAI
Moonshot AI
Anthropic
Google
Google
OpenAI
OpenAI
xAI
Thinking Machines
Thinking Machines
Meta
Meta
Alibaba
Alibaba
68.9%
🥇AnthropicAnthropicClaude Opus 572.1%
🥈AnthropicAnthropicClaude Sonnet 570.0%
🥉OpenAIOpenAIGPT-5.6 Sol67.3%
4xAIxAIGrok 4.664.2%
5Moonshot AIMoonshot AIKimi K3OPEN66.3%
6AnthropicAnthropicClaude Fable 568.9%
7GoogleGoogleGemini 3.5 Flash Lite68.7%
8GoogleGoogleGemini 3.6 Flash67.6%
9OpenAIOpenAIGPT-5.6 Luna67.3%
10OpenAIOpenAIGPT-5.6 Terra67.3%
11xAIxAIGrok 4.561.8%
12Thinking MachinesThinking MachinesInklingOPEN64.0%
13Thinking MachinesThinking MachinesInkling SmallOPEN62.3%
14MetaMetaMuse Spark 1.166.1%
15MetaMetaMuse Spark 1.265.4%
16AlibabaAlibabaQwen 3.7 Plus66.2%
17AlibabaAlibabaQwen 3.8 MaxOPEN64.0%
Full Results Table
#ModelOverallSourceDate
🥇
AnthropicClaude Opus 5
Anthropic
72.1VALS (2026)2026-07
🥈
AnthropicClaude Sonnet 5
Anthropic
70.0VALS (2026)2026-06
🥉
AnthropicClaude Fable 5
Anthropic
68.9VALS (2026)2026-06
4
GoogleGemini 3.5 Flash Lite
Google
68.7VALS (2026)2026-07
5
GoogleGemini 3.6 Flash
Google
67.6VALS (2026)2026-07
6
OpenAIGPT-5.6 Terra
OpenAI
67.3VALS (2026)2026-07
7
OpenAIGPT-5.6 Sol
OpenAI
67.3VALS (2026)2026-07
8
OpenAIGPT-5.6 Luna
OpenAI
67.3VALS (2026)2026-07
9
Moonshot AIKimi K3OPEN
Moonshot AI
66.3VALS (2026)2026-07
10
AlibabaQwen 3.7 Plus
Alibaba
66.2VALS (2026)2026-06
11
MetaMuse Spark 1.1
Meta
66.1VALS (2026)2026-07
12
MetaMuse Spark 1.2
Meta
65.4VALS (2026)2026-08
13
xAIGrok 4.6
xAI
64.2VALS (2026)2026-08
14
Thinking MachinesInklingOPEN
Thinking Machines
64.0VALS (2026)2026-07
15
AlibabaQwen 3.8 MaxOPEN
Alibaba
64.0VALS (2026)2026-08
16
Thinking MachinesInkling SmallOPEN
Thinking Machines
62.3VALS (2026)2026-07
17
xAIGrok 4.5
xAI
61.8VALS (2026)2026-07
All evaluated models
Anthropic
Claude Opus 5
72.1%
Anthropic
Claude Sonnet 5
70.0%
OpenAI
GPT-5.6 Sol
67.3%
xAI
Grok 4.6
64.2%
Moonshot AI
Kimi K3
66.3%
Anthropic
Claude Fable 5
68.9%
Google
Gemini 3.5 Flash Lite
68.7%
Google
Gemini 3.6 Flash
67.6%
OpenAI
GPT-5.6 Luna
67.3%
OpenAI
GPT-5.6 Terra
67.3%
xAI
Grok 4.5
61.8%
Thinking Machines
Inkling
64.0%
Thinking Machines
Inkling Small
62.3%
Meta
Muse Spark 1.1
66.1%
Meta
Muse Spark 1.2
65.4%
Alibaba
Qwen 3.7 Plus
66.2%
Alibaba
Qwen 3.8 Max
64.0%