Ptally vs Akur8 Deploy
How Ptally compares with Akur8 Deploy, Akur8's cloud rating engine: where each fits, how rates get into production, testing and approvals, AI features and commercial model.
If your actuaries already build rating plans in Akur8 and you want those plans live with as few handovers as possible, Akur8 Deploy is the natural choice. If you want a rating engine that starts from any source (a rating guide, an Excel rater or a US rate filing), covers the full product rules as well as the rates, and has no upfront cost, Ptally is the better fit. The two can also work together: Akur8 to build the rating structure from data, Ptally to implement, test and run it.
What Akur8 Deploy is
Akur8 is best known for its pricing modelling software, which it describes as "powered by transparent machine learning". Its modules include Data, Risk, Demand and Rate. In October 2025 Akur8 added two products: Rate Repo, a rate repository with versioning and documentation, and Deploy, a cloud rating engine for putting Akur8 rates into production.
Akur8 says Deploy lets teams "import rate plans in minutes, deploy them in seconds, and quote in milliseconds". It exposes rates through a real-time API that connects to a policy administration system, and supports testing before go-live, A/B tests, simulations and routing strategies that segment the client base. Every change is versioned and logged, and teams can review and approve changes that affect production. In August 2026 Akur8 released a Guidewire Marketplace accelerator so PolicyCenter customers can call Deploy for rating without a PolicyCenter redeployment.
Akur8 reports more than 350 customers in over 40 countries and over 3,000 actuaries using its platform daily. In October 2026 it launched Akur8 Agents, AI agents that work inside the Akur8 platform to support actuarial workflows, with actuaries making the final calls.
What Ptally is
Ptally is an AI enabled insurance pricing platform for insurers and MGAs. A model is a readable JSON file of ordered steps: lookups, calculations, transforms, exclusions and referrals. Every change runs against the model's regression test suite, goes through draft, approval and publishing, and is kept in the version history. The same engine serves a real-time API, embedded quote forms, and chat and voice agents, and has served over 50 million quotes, including at aggregator volume.
AI is built into the workflow. It extracts rate tables from your documents and drafts the model, and Claude, ChatGPT or Cursor can connect to Ptally over MCP to read, validate, test and edit draft models. Agents can only change drafts. Approving and publishing stay with people.
Side by side
| Ptally | Akur8 Deploy | |
|---|---|---|
| What it is | AI enabled pricing platform and rating engine | Cloud rating engine within the Akur8 pricing suite |
| Main source of rates | Rating guides, Excel raters, US rate filings or rates from any modelling tool | Rating plans built in Akur8 |
| Model format | Readable JSON of ordered steps, tests included | Not published in detail |
| Product rules | Exclusions, referrals, excesses and endorsements in the same model | Public material focuses on rates, simulations and routing |
| Testing | Regression test suite runs on every change | Pre go-live testing, simulations and A/B tests |
| Approvals and audit | Draft, approve, publish; full version history; every quote tied to a version | Versioned changes, audit logs, review and approval |
| AI | AI drafts and checks models; MCP server for Claude, ChatGPT and Cursor | Akur8 Agents in the wider platform; machine learning in modelling |
| Integration | Real-time API, embedded forms, chat and voice | Real-time API; Guidewire PolicyCenter accelerator |
| Commercial model | Free to build. Up to 1% of GWP, a flat fee per model, or Enterprise terms | Not published |
The key differences
Modelling versus implementation
Akur8's strength is building the rating structure from data. Its GLM and machine learning tools are widely used by pricing actuaries, and Deploy shortens the path from that work to a live price. Ptally does not fit models from claims data. It takes a rating structure that has already been designed, wherever it came from, and turns it into a tested, versioned model that prices live quotes.
What counts as the model
Deploy is built around rating plans from Akur8. A live product usually needs more than rates: decline and referral rules, excesses, fees, taxes, rounding and policy wording. In Ptally all of these sit in one model as ordered steps, and endorsement steps can drive the policy documents built from each quote.
Testing and traceability
Both products version changes and support review before release. Ptally's approach is a regression test suite in every model. Each test is a full set of inputs with the expected result, and the suite runs on every change, so a rate change shows exactly which quotes moved. Every quote response records the version that priced it, and the debug trace shows every input, lookup and formula.
How AI is used
Akur8 applies machine learning to modelling and has added agents that guide actuaries inside its platform. Ptally applies AI to building and maintaining the production model. It reads your documents, drafts the steps and checks them against your worked examples, and its MCP server lets general AI clients work on drafts under the same tests and approvals as people.
Cost
Akur8 does not publish pricing. Ptally is free to build, with no implementation fee. On the Performance plan you pay up to 1% of the gross written premium your models write, so the first cost you see is a share of premium you are already writing.
When Akur8 Deploy is the better choice
- Your pricing team already models in Akur8 and wants rates to move from model to production inside one suite.
- You want built-in A/B testing and routing strategies on live rates.
- You run Guidewire PolicyCenter and want a vendor accelerator for external rating.
- Your policy system handles the product rules and you only need an external engine for rates.
When Ptally is the better choice
- Your rates live in a rating guide, a spreadsheet or a filing, not in a modelling tool.
- You want rates and product rules in one readable, tested model.
- You want to launch without a licence or implementation project, and pay as premium is written.
- You want AI clients such as Claude or ChatGPT to work on models, with people approving every release.
- You need to quote at aggregator volume through an API, forms or chat.
If you have no in-house capacity to build the model, Ptally's pricing experts can build it for you from your documentation, with no implementation fee.
Using both
Many teams will keep Akur8 for modelling. The rates and relativities that come out of Akur8 can be uploaded to a Ptally model as CSV rate tables, with the rest of the product rules around them and tests on every change.
Next steps
See how a model is built and tested on the developer page, how quotes are served through the API, or compare the pricing plans. For a wider view of the market, read Ptally compared with established platforms and new AI entrants.
This comparison reflects publicly available information as of November 2026. Vendor offerings change, so check current details with each vendor.