Best AI model for finance and accounting — November 2026
No single model is clearly best for finance and accounting on BenchLM's evidence. On Vals Finance Index, run by Vals AI, Gemini 4 Carbon (67.0%), Gemini 4 Hydrogen (66.1%) and Gemini 4 Argon (66.0%) are within 1.5 points of each other. A benchmark result is not an accounting or tax qualification. Figures, filings and advice need review by a qualified person, and the index tests none of your own ledgers or software.
The shortlist for finance and accounting
Constraints on this page: a builder choosing by accuracy, hosted processing allowed, any price, ordinary input size. Change any of them under Refine. A small score gap does not establish a reliably better model.
Compare Gemini 4 Carbon vs Gemini 4 HydrogenAll Vals Finance Index results
01Gemini 4 CarbonTied at the top
Google · Proprietary
67.0%
Vals Finance Index
- Highest Vals Finance Index result, with 2 other models within 1.5 points
- $4.00 for the stated workload
- 1M context
- Google lists gemini-4-carbon as a stable model, generally available on the paid tier since November 10, 2026, after a limited first release through the Fairwind Program. Access details · checked November 10, 2026.
- Provider API offer: Gemini API (Google AI Studio). Check which models qualify and whether payment details are required. Limits and source.
Tested setting: Gemini 4 Carbon (high effort)
02Gemini 4 HydrogenTied at the top
Google · Proprietary
66.1%
Vals Finance Index
- 66.1% on Vals Finance Index, run by Vals AI
- $3.00 for the stated workload
- 1M context
- Google lists gemini-4-hydrogen as generally available on the free and paid tiers since November 12, 2026. Access details · checked November 12, 2026.
- Provider API offer: Gemini API (Google AI Studio). Check which models qualify and whether payment details are required. Limits and source.
Tested setting: Gemini 4 Hydrogen (xhigh effort)
Model page and evidenceCompare with Gemini 4 CarbonOpen benchmark source
03Gemini 4 ArgonTied at the top
Google · Proprietary
66.0%
Vals Finance Index
- 66.0% on Vals Finance Index, run by Vals AI
- $4.00 for the stated workload
- 1M context
- Google lists gemini-4-argon as a stable model. Access details · checked October 13, 2026.
- Provider API offer: Gemini API (Google AI Studio). Check which models qualify and whether payment details are required. Limits and source.
Tested setting: Gemini 4 Argon (high effort)
Model page and evidenceCompare with Gemini 4 CarbonOpen benchmark source
04Claude Opus 5.5
Anthropic · Proprietary
65.3%
Vals Finance Index
- 65.3% on Vals Finance Index, run by Vals AI
- $8.00 for the stated workload
- 1M context
- Anthropic lists the model ID on each documented platform. Access details · checked September 22, 2026.
- Provider API offer: Claude API. Check which models qualify and whether payment details are required. Limits and source.
Tested setting: Claude Opus 5.5 (max effort)
Model page and evidenceCompare with Gemini 4 CarbonOpen benchmark source
05Muse Wonder Full
Meta · Proprietary
64.6%
Vals Finance Index
- 64.6% on Vals Finance Index, run by Vals AI
- 1M context
- API access is not verified. Check the provider documentation before relying on it.
Tested setting: Muse Wonder Full (max effort)
Model page and evidenceCompare with Gemini 4 CarbonOpen benchmark source
ChatGPT, Claude, Gemini, or Copilot?
Of the three apps whose models we can score, Gemini has the highest result here: Gemini 4 Carbon scores 67.0% on Vals Finance Index.
- ChatGPTGPT-6.1 Sol scores 55.8%Plus, $20 a monthPlan lists: GPT-6 Astra, GPT-6.1 Sol, GPT-6.1 Terra, GPT-6 Sol and GPT-6 Luna in Codex; GPT-6.1 Sol is available in Work and Codex, not Chat.
- ClaudeClaude Opus 5.5 scores 65.3%Pro, $20 a monthPlan lists: Opus, Sonnet and Haiku; Fable is listed under usage credits rather than the included allowance.
- GeminiGemini 4 Carbon scores 67.0% (highest of the three)Google AI Pro, $19.99 a monthPlan lists: Gemini 4 Argon, the Pro model and the Flash model.
- Microsoft CopilotNot scored. Microsoft doesn’t name the model Copilot runs, so no benchmark result can be attached to it. How it compares on plans and data termsMicrosoft 365 Premium, $19.99 a month
Each entry is the highest-scoring model that company sells, not the model your plan opens by default, so check what the plan lists before you pay. The plan shown is each app’s plan nearest $20, read October 16, 2026 for ChatGPT, November 10, 2026 for Copilot, and November 11, 2026 for Claude and Gemini.
ChatGPT vs ClaudeChatGPT vs GeminiClaude vs GeminiChatGPT vs Copilot
The top ten for finance and accounting, with API prices
The first ten models on the same evidence and constraints as the shortlist, with each provider's API price in US dollars per million tokens. Inkling 2 Small costs the least of the ten for a workload of 1M input and 200k output tokens: $0.30 per million input tokens and $1.00 per million output tokens. Two of the ten have no API price on file.
| # | Model | Vals Finance Index | Input | Output |
|---|---|---|---|---|
| 01 | Gemini 4 CarbonGoogle | 67.0% | $2.00 | $10.00 |
| 02 | Gemini 4 HydrogenGoogle | 66.1% | $1.50 | $7.50 |
| 03 | Gemini 4 ArgonGoogle | 66.0% | $2.00 | $10.00 |
| 04 | Claude Opus 5.5Anthropic | 65.3% | $4.00 | $20.00 |
| 05 | Muse Wonder FullMeta | 64.6% | No API price on file | |
| 06 | Claude Sonnet 5.5Anthropic | 64.5% | $2.00 | $10.00 |
| 07 | Claude Fable 5.2Anthropic | 63.6% | $8.00 | $40.00 |
| 08 | Inkling 2 SmallThinking Machines Lab | 61.5% | $0.30 | $1.00 |
| 09 | GLM-5.3Z.AI | 61.4% | No API price on file | |
| 10 | Muse WonderMeta | 61.2% | $4.00 | $20.00 |
Prices are list API rates before caching, batch discounts and special tiers. Compare API pricing for every model.
Refine for your situation
Each link opens the selector with one answer changed. The address carries the answers, so your version is as shareable as this page.
- Change any answer in the selector
- Must run on my own hardware
- Cost matters most
- Speed matters most
- Long documents or a large codebase
- Choosing an everyday assistant, not an API
- Describe your own job instead
We keep what you type here, drafts included, for 180 days, and after that only the job and requirements read from it. Nothing is tied to your IP address or account. Leave out names and other personal details. Privacy
What to verify before choosing
- A benchmark result is not an accounting or tax qualification. Figures, filings and advice need review by a qualified person, and the index tests none of your own ledgers or software.
- Ranked on one benchmark: Vals Finance Index, run by Vals AI. 36 models have a published result; a model without one is not listed.
Try three representative examples of your own work. Compare errors, time, cost, and the tools available in your actual setup.
Before your company approves a model
The shortlist ranks models. Whether you can use one at work depends on the provider: whether it trains on what you send, how long it keeps it, and what it will sign. The table covers the providers behind the models above, plus OpenAI, and the two clouds that sell many of the same models under their own terms. Each cell is what the provider’s own pages said on November 10, 2026.
With ledgers, payroll, or client tax files, we’d read the retention and SOC 2 columns first.
| Provider | Trains on your inputs? | Inputs kept | SOC 2 report | HIPAA | Single sign-on |
|---|---|---|---|---|---|
| GoogleGemini API, Gemini on Google Cloud, and Gemini in Google Workspace | Trains on your inputs?Not on paid plans | Inputs keptNo number given | SOC 2 reportType II | HIPAABAA: Cloud and Workspace | Single sign-onThrough Workspace |
| AnthropicClaude API, Claude Team, and Claude Enterprise | Trains on your inputs?No, by default | Inputs kept30 days, with exceptions | SOC 2 reportType II | HIPAABAA: API and Enterprise | Single sign-onTeam and Enterprise |
| OpenAIOpenAI API, ChatGPT Business, and ChatGPT Enterprise | Trains on your inputs?No, by default | Inputs kept30 days; state until deleted | SOC 2 reportType 2 | HIPAABAA on request, API | Single sign-onBusiness, Enterprise, Edu |
| Microsoft Foundry (Azure)Cloud. Models sold by Azure, including OpenAI, xAI, Meta, Mistral, and DeepSeek models | Trains on your inputs?No | Inputs keptFlagged prompts may be kept | SOC 2 reportType 2 | HIPAABAA, in-scope services | Single sign-onMicrosoft Entra ID |
| Amazon BedrockCloud. Models from Anthropic, OpenAI, Meta, Mistral, xAI, DeepSeek, Moonshot AI, Qwen, Z.AI, and others | Trains on your inputs?No | Inputs keptNo period given | SOC 2 reportIn audit scope | HIPAAEligible, with a BAA | Single sign-onIAM Identity Center |
- “Not stated” means the pages we read don’t say. It doesn’t mean the provider lacks it.
- Each row covers the plans named under the provider. A personal plan has its own terms, and your signed agreement overrides the public page.
- Meta: We found no page that sets out business data terms for Meta’s Model API. Its models are also sold on Microsoft Foundry (Azure), Amazon Bedrock, and Google Cloud (preview), where that cloud’s terms apply.
- Z.AI: Z.AI’s developer privacy policy does not say plainly whether API inputs are used for training. Its models are also sold on Amazon Bedrock and Google Cloud, where that cloud’s terms apply.
- We haven’t read the terms for Thinking Machines Lab yet.
What this shortlist rests on
One benchmark: Vals Finance Index. Agent work a finance team does: analyst research on company filings, building financial models in Excel, and corporate tax research. Vals AI combines the three into one index. A result a model's own provider ran is left out, and a model nobody has run there is not listed.
- Vals Finance IndexSets the order. Run by Vals AI. 36 models have a published result.
- APEX-Accounting (Mercor)Related evidence, not used for the order. Month-end close tasks across files and applications, run by Mercor on a few agent setups.
- TaxEval v2Related evidence, not used for the order. Tax questions answered without tools, also run by Vals AI on more models.
- CorpFin v2Related evidence, not used for the order. Questions over long credit agreements, also run by Vals AI.
Questions
Which AI model is best for finance and accounting?
No single model is clearly best for finance and accounting on BenchLM's evidence. On Vals Finance Index, run by Vals AI, Gemini 4 Carbon (67.0%), Gemini 4 Hydrogen (66.1%) and Gemini 4 Argon (66.0%) are within 1.5 points of each other. A benchmark result is not an accounting or tax qualification. Figures, filings and advice need review by a qualified person, and the index tests none of your own ledgers or software.
Which models come next for finance and accounting?
Claude Opus 5.5 (65.3%), Muse Wonder Full (64.6%) follow on the same benchmark. Compare the leaders on three representative examples of your own work.
How does BenchLM pick the best ai model for finance and accounting?
The page runs the LLM Selector with fixed answers: a builder choosing by accuracy, hosted processing allowed, any price, ordinary input size. The selector uses Vals Finance Index, one benchmark run by Vals AI on every listed model, filters by the stated constraints, and orders by that evidence. It never adds a hidden fit score or a bonus for open weights or reasoning style.
Why does this page use Vals Finance Index and not a general score?
A general score answers several jobs at once. Vals Finance Index is closer to this job. Agent work a finance team does: analyst research on company filings, building financial models in Excel, and corporate tax research. Vals AI combines the three into one index. Only results Vals AI published itself are used, so a model nobody has run there is not listed. 36 models have a result.
What do the top models for finance and accounting cost?
Among the ten highest-ranked models on this page, Inkling 2 Small costs the least for a workload of 1M input and 200k output tokens: $0.30 per million input tokens and $1.00 per million output tokens. Claude Fable 5.2 costs the most, at $8.00 and $40.00. These are API rates in the November 13, 2026 data. Assistant subscriptions are priced separately.
Can I change the constraints?
Yes. Every link under "Refine" opens the selector with one answer changed, and the address carries the answers so a result can be shared or reopened against the current dataset.
Method: bench-align-v5.8-2026-09-30. Read the methodology and benchmark confidence pages for how scores and verification statuses are produced.
Watch the finance and accounting shortlist
One weekly email when rank, price, or benchmark evidence changes make this shortlist worth revisiting.
Read a sample issueJoin 7,000+ subscribers.