Lokta vs OpenCBS*

Open source lets you read the core. Lokta lets you audit what runs on the book.

Source you can read is not source you are running. What you have to prove to a regulator is what ran on your book. OpenCBS is an open-source core for microfinance and digital lenders, forked from Octopus Microfinance Suite. We run the loan book after approval on a commercial licence, deployable on infrastructure you operate, and a servicing action leaves a record: the proposal, the decision, the policy version and the approver.

  1. Agent proposesThe next move for one account, with the reason attached.
  2. Core decidesPosts it, or refuses it, against the policy version in force.
  3. Record proves itProposal, decision, policy version and approver, on the account.
How every agent action on Lokta runs. AI proposes, the core decides, the record proves it.

Choose Lokta if

Your core is settled, open source or otherwise, and you need to prove what ran on the book after approval, on infrastructure your own team operates.

Look at both if

OpenCBS stays the open-source core of record and Lokta runs as the commercial post-approval layer above it, on infrastructure you operate.

Why choose Lokta for the book after approval?

Depth where the book gets hard

Lokta puts its depth in one place: delinquency, collections strategy, treatment allocation, promise-to-pay tracking and recovery on a live book. It is built by the team behind Apache Fineract and deployed by engineers who work alongside yours, in your own tree, on infrastructure you operate. Agents test servicing and collections strategies continuously and propose the winners; the deterministic core decides whether each is inside policy before anything posts. OpenCBS gives a small institution a working core with the source that comes with it, and we are not the cheaper option.

We are the team behind Apache Fineract, the open-source lending core in production around the world, and we went on to build and scale a commercial lending platform on those foundations. More on the team. We do the book after approval and are not building origination this cycle.

Before you shortlist anyoneShould you move at all?A comparison tells you how two platforms differ. It does not tell you whether moving is worth it for your book. The migration assessment works through that in six stages, against your own deployment, and it can end with a recommendation to stay where you are.

When is OpenCBS still the right choice?

  • You want the source under a copyleft licence and you have the engineering capacity to own it. Source for OpenCBS Cloud and its loan origination repository is published on GitHub under GPL-3.0, on a substantial real codebase. That is a right our commercial licence does not give you.
  • You need a complete core banking system, with origination available as a separate OpenCBS product: teller and vault management, savings, term deposits, provisioning, fixed assets and end-of-day processing. We do none of it.
  • You are a small institution where OpenCBS states an affordable and flexible pricing structure fits. We are not priced for that institution and do not try to be.

Neither replaces the other. OpenCBS stays the system of record with the source you can read; Lokta runs above it on infrastructure you operate, and the record of every servicing action lives on our side of that line.

The agents cover the whole book

The agents are the fabric of the platform. Each one watches the live book, proposes the next move, and hands it to the deterministic core to post or refuse. You get the work done; the core keeps the authority.

  • Monitoring and analyticsEvery account watched continuously, not sampled at month-end.
  • Early warningThe account about to slip is flagged before it does, with the reason attached.
  • Next best action for collectionsFor each account in arrears, the treatment most likely to cure it, ranked and evidence-backed.
  • Servicing requests, non-voiceStatements, payoff requests and schedule changes drafted and routed, never posted without the core.
  • RecoveryThe late book worked by strategy, not by whoever has capacity that week.
Deterministic corePosts the move or refuses it. The agent never writes to the ledger.Every decision on the record

What does one governed action leave behind?

Every platform here can write the sentence. The record is the part that has to be true, so this is the artifact rather than the claim. The products behind it are composed the same governed way in the Loan Product Studio.

What one governed action leaves behind

Agent proposes
Hold 1 account from the dialler for 7 days, promise to pay recorded
Policy bound
v14, promise-to-pay hold max 7 days, 31-90 days past due
Core decides
Inside policy. Hold posted to the account.
Record
Proposal, decision, policy version and approver, on the account
The agent proposes. It does not post to the ledger. The core decides, and the record carries why. That is what the matrix below measures.

Why do lenders look for an OpenCBS alternative?

  1. You cannot tell whether the hosted code is the published code.

    The licence file on OpenCBS Cloud is the complete, unmodified GPL-3.0, not AGPL-3.0, so a hosted modification to the code running your accounts carries no obligation to be published.

  2. You do not know when the next security patch lands.

    OpenCBS Cloud carries four commits, no releases and no tags, with no code pushed since January 2023, and the desktop product is end-of-life per its SourceForge listing.

  3. After approval you get delinquency reporting, not a strategy layer.

    OpenCBS describes delinquency flagging, action tracking and automated warning letters inside the core; a strategy layer with treatment testing, agent allocation and promise-to-pay tracking is not described in its public material.

How do Lokta and OpenCBS compare, capability by capability?

Capability
OpenCBS
Lokta
Source code published under a copyleft licencePublished on GitHub under GPL-3.0 for OpenCBS Cloud and its origination repository. Our commercial licence gives you no such right.
Yes
No
Full core banking including teller, vault, savings and term depositsWe do none of it. If you take deposits, this comparison ends here.
Yes
No
Origination and servicing from one vendorOrigination is a separate OpenCBS product, the Loan Origination Solution, which carries the automated decision engine; their functional-coverage table marks it absent from OpenCBS Cloud itself. We are not building origination this cycle.
Yes
No
Collections strategy testing, treatment allocation and recovery workflowTheir published depth after approval is delinquency management inside the core; a treatment-testing layer is not described.
Not stated
Yes
Agents operating the book inside policy boundsNot described in OpenCBS public material. Their automation digitises the lending process for people to run.
Not stated
Yes
Published release and security-patch cadence for the hosted productThe public repositories are quiet and hosted changes need not be published. Ask us for ours too.
Not stated
No
Maker-checker and audit trail as named controlsBoth name them as first-class controls. The difference is whether the control also covers what a model proposed.
Yes
Yes

Yes and No mean the capability is described, or ruled out, in public material. Via partner means the vendor supplies it through a named partner rather than in the product. Not stated means we could not find it in OpenCBS’s published material, which is not the same as absent from the product. We do not use that state in our own column: about ourselves the answer is Yes or No. Rows where the answer goes against us are here on purpose.

What is the difference between Lokta and OpenCBS?

What the open licence gets you

OpenCBSGPL-3.0 source published for OpenCBS Cloud and its origination repository, delivered as hosted SaaS on AWS, with custom hosting available and customisation scoped per client.
LoktaA commercial licence, deployed on-premises, in single-tenant cloud or inside your own VPC, with extension in your own tree instead of change requests against someone else’s console.

If the requirement is owning the source, choose them and stop here. If it is knowing what runs on your accounts, ask for the commit that matches your instance; the licence does not make them publish it.

Lineage

OpenCBSA fork of the last open-source release of Octopus Microfinance Suite, a 2006-origin codebase on a different technology stack.
LoktaBuilt by the team behind Apache Fineract, on a new codebase, not a distribution of it, designed around a core whose job is to adjudicate what an agent proposes.

OpenCBS and Fineract share no code and no history. Ask the same question of either codebase: was the boundary between a model and the ledger there from the first commit, or added to a core built for people to run?

Depth after approval

OpenCBSDelinquency management inside a core banking system: flagging, action tracking and automated warning letters, alongside accounting with provisioning and write-offs.
LoktaCollections strategy, treatment allocation, agent assignment, promise-to-pay tracking and recovery workflow as the product, on a live book. Agents test hundreds of collection strategies in the time a risk team ships one.

Delinquency reporting tells you which accounts are bad. A strategy layer decides what to do about each one, tests whether it worked, and promotes the version that did.

Operating model

OpenCBSDigitises the lending process for people to run, with workflow automation, an origination decision engine and a digitised credit committee.
LoktaAgents run the volume inside policy bounds, people own the exceptions, and the deterministic core owns the arithmetic: the ledger and money-of-record are computed by code, never by a model.

One model makes your team faster. The other changes how many people the book needs at all, which is the lever that still works when the book doubles.

Governance controls

OpenCBSMaker-checker and audit trails named as first-class controls, the right instinct for a regulated lender.
LoktaThe same two controls, with the deterministic core in the decision path: an agent-proposed change is adjudicated against the current policy version before it posts, and the approver is named on each one.

Maker-checker is not our differentiator. What to test is whether the control covers a posting a model proposed, or only one a person entered, and where in the code or the record you can see it.

Scope

OpenCBSCore banking for microfinance: client management, loan management, savings, teller and vault, accounting and analytics, with origination and its decision engine sold as a separate Loan Origination Solution product.
LoktaPost-approval only, for NBFCs and fintech lenders, on your core or on ours: servicing, monitoring and early warning, collections and recovery. We are not building origination this cycle.

If you need a system of record with its own ledger, we are not a candidate. The comparison is real for an institution that already has a core and wants to change how the loan book is run above it.

Where would we expect an argument?

Reading the source is not the same as governing what runs on it. After approval the control that matters is whether every agent action is checked against policy and written to the record before it posts, and no repository gives you that for free.

What should you ask OpenCBS in an evaluation?

  • Is the code running in the hosted product the code published on GitHub, commit for commit? Under GPL-3.0 that is a question about verifiability, not about conduct.
  • What is the release and security-patch cadence for the hosted product, and where is it published? The public repositories have had no code pushed since January 2023.
  • Is the tablet application source available? We could not find a public repository for it.
  • When automation acts on an account, in either the open-source or the hosted product, what checks the change against policy before the posting, and does the record carry the rule it satisfied?

Each is a diligence question we could not settle from OpenCBS’s public material as of August 2026, not a claim about what the product does. Ask them, and ask us the equivalent. Our AI governance questions and the full evaluation questionnaire are the set we would put to any platform, including ours.

In lending, the autonomy you can audit is the only autonomy that scales.

Start a conversation

Lokta is the agentic loan servicing platform: we run the book after approval. If your book is already disbursed and the cost of running it is growing with it, that is the problem we are built for. If it is not, one of the sections above will tell you so faster than a demo will.

Frequently asked questions about OpenCBS

Is Lokta an OpenCBS alternative?

For a buyer who wants a system of record, no. OpenCBS is a core banking system for microfinance institutions, financial cooperatives and digital lenders, with source published on GitHub. Lokta runs the loan book after approval for NBFCs and fintech lenders, on your core or on ours. The comparison is real for one reader: the lender whose core is settled and whose cost and delinquency now sit in servicing, collections and recovery, on accounts the core only reports.

Is OpenCBS related to Apache Fineract or Mifos?

No. OpenCBS began as a fork of the last open-source release of Octopus Microfinance Suite, a project that started in 2006, and runs on a different technology stack. It has no Mifos or Apache Fineract relationship at any point. Lokta’s team helped build Apache Fineract and then started a new codebase, so the two lineages on this page never touched.

Is OpenCBS open source?

Partly, and the distinction is worth an afternoon. Source for OpenCBS Cloud and its loan origination repository is published on GitHub under GPL-3.0. The desktop product is end-of-life per its SourceForge listing and carries no licence file. Because the licence is GPL-3.0 and not AGPL-3.0, a modification made in the hosted product carries no obligation to be published, so the code you can read and the code serving your accounts can differ. Ask for the commit.

Lokta is not open source. Why choose it over something we can own?

You are buying the control, not the code. If owning the source is the requirement, choose OpenCBS. Choose Lokta if the requirement is that the book after approval is run inside a policy you set, with proof: agents propose the treatment for each account in arrears, the core refuses anything outside policy, a person approves every promoted strategy, and the record of all of it is yours to hand a regulator.

How does Lokta prove what ran on our book?

By keeping the decision on the record, beside the outcome. The agent proposes an action on an account. It does not post to the ledger. The deterministic core decides whether the proposal is inside the current policy version, and the record carries the proposal, the decision, the policy version and the approver. Anything outside policy goes to a named approver under maker-checker, and a refusal is recorded the same way.

Can Lokta run on our own infrastructure alongside OpenCBS?

Yes. Lokta deploys on-premises, in single-tenant cloud, or inside your own VPC, and it is the same binary across all three, so the platform can sit on infrastructure your team operates while OpenCBS remains the system of record. OpenCBS itself is delivered as SaaS on AWS, with other hosting referenced under custom offers, so put the deployment question to them directly. Lokta engineers deploy alongside your team, and extension lives in your own tree. Nothing here moves your ledger.

Sources and method

Everything on this page about OpenCBS is drawn from its own public material, linked below and last checked on 20 August 2026. We do not run hands-on testing of other platforms, so a row records what the vendor states rather than what we have verified in a deployment, and where the material does not settle a question the row says so instead of treating silence as absence. Capabilities change and deployments differ: confirm the current position, and your own configuration, with the vendor.

Founder-led adoption

Adopt the agentic loan servicing platform.

Lokta is built for enterprise deployment, VPC or single-tenant cloud, with an audit trail in every state change. We work with a select group of institutions through a founder-led model: deep adoption, deliberate scope, a delivery window the team commits to in writing.