Skip to content
An independent field guide to AI character platformsResearch edition · October 2026
charactersFind your fit
A SOURCE-BASED COMPARISON

Kindroid vs NomiThe details behind the choice.

Start with Kindroid if you want an explicit Response Directive and editable response examples. Consider Nomi if you prefer initial traits with background and temporary roleplay context in Backstory+. Both document group conversations. The differences below explain what you can configure; conversation quality and recall have not been benchmarked.

Guide revised 28 Oct 2026Sources reviewed 3 Oct 2026Documentation review · products not hands-on tested

Documented differences. Editorial interpretation. The starting points below follow from public controls and offers. We have not measured conversation quality, memory recall or paid-plan performance.

Start with the job you need it to do.

Kindroid

For explicit response instructions

Kindroid separates Personality (formerly Backstory), a Response Directive and an Example Message. This gives a writer distinct places for identity, a concise behavior instruction and a sample of the desired voice.

Read the platform passport
Nomi

For a trait-led starting point

Nomi starts with selected traits and interests. Nomipedia says those starting choices cannot be edited directly afterward; Backstory+ adds context you can revise as the character and scene develop.

Read the platform passport

What the documentation establishes.

Known controls retain their plan and setup qualifications. “Not established” is a gap in our evidence, rather than proof a feature is absent.

Scroll sideways to compare both platforms.

Setup differences · sources checked 28 Oct 2026
ControlKindroidNomi
Changing the personality setup

Edit Personality, Response Directive and Example Message as separate controls.

Kindroid documentation

Initial traits stay fixed; revise context through Backstory+.

Nomi documentation
Keeping a lasting fact

Key Memories and journals are distinct from automatically updated Learned Context. Access depends on the tier or available preview.

Kindroid documentation

Backstory holds enduring context; Current Roleplay holds temporary scene details.

Nomi documentation
Directing a group scene

Paid groups, up to ten Kindroids; shared context, Scene and manual or automatic turns.

Kindroid documentation

Group Backstory and a separate Mind Map; select the next speaker or use Auto-play.

Nomi documentation

How much do you want to write yourself?

Kindroid’s current guide calls its main background field Personality; older references call it Backstory. The Response Directive supplies concise behavioral guidance, while an Example Message demonstrates how to speak. Nomi’s current wiki distinguishes lasting Backstory from temporary Current Roleplay, both within Backstory+.

Original example to adapt: a museum guide named Jun is patient but skeptical of an unsigned map. Keep “Jun asks one question before drawing a conclusion” as an enduring behavior. Keep “The museum closes in twenty minutes” as a scene condition. When the museum opens tomorrow, update the scene condition instead of rewriting the character’s identity.

Which gives you the memory controls you need?

Kindroid documents editable Learned Context alongside Key Memories and journals. Nomi describes several memory layers plus Backstory+, Identity Core and Mind Maps. Those are different ways to supply or organize context, not comparable measurements of recall accuracy.

For a useful check, record one invented fact in a reference field and introduce a different fact only through dialogue. Ask about each later, then correct one of them. Record the plan, elapsed time, whether the reply needed a hint and whether the correction persisted. Keep written reference and conversational recall as separate results.

A shared scene needs its own context

Kindroid’s group guide describes shared context, a Group Directive and a Scene field. Nomi’s group guide describes a group Backstory and a group Mind Map. Both let you choose the next speaker or use automatic turns. Put shared facts in the group’s context: neither guide says participants have full access to one another’s individual backgrounds.

For Jun’s museum scene, give a second adult character a different goal: preserve the map while Jun wants to identify it. Put the closing time and the map’s location in the shared scene. Check whether each speaker keeps a distinct position before adding more participants. A higher participant limit alone cannot establish a better group conversation.

Check the trial conditions before comparing costs

Kindroid’s current subscription guide describes a three-day trial requiring a payment method that converts to a subscription unless cancelled. Its separate new-account preview is a one-time allowance, not an ongoing daily free tier. The guide directs users to checkout for the price applicable to their country and platform.

Standard, Ultra and MAX are not interchangeable price labels: the guide describes the latter two as additional monthly subscriptions with dependencies. Compare the full combination you would pay for. This review has not established comparable current cash prices for both services, so it does not name a cheaper platform.

Voice and API access are separate decisions

Both platforms have official API documentation for interacting with existing characters. Kindroid also documents group endpoints that require a subscription. Nomi describes account keys for integrations. Atlas has not created keys or sent requests to either API.

Both document voice features, but that does not establish the same allowance, latency or quality. Compare account access and model behavior before designing an integration around them. Character export and local installation remain unverified in this comparison.

Costs & allowances.

Current subscription amounts and renewals for these two platforms have not been established in Atlas. Paid features are identified where the documentation says so; no cheapest-plan claim or numerical cost estimate is assigned.

A useful first test.

Suggested exercise, not a test we performed. Two adult expedition partners disagree about which route to take. One favors caution, the other speed. Both know that a bridge has closed.

  1. Keep the motives and initial background the same in both products.
  2. Observe whether each participant keeps a distinct position and speaking style.
  3. Return later and ask about the closed bridge; record the access tier and elapsed time.
  4. Try one deliberate personality edit and note which settings you could change.
  5. Record written reference recall and dialogue-only recall separately; repeat after correcting one fact.
Record the comparison in the memory worksheet

Common questions.

Should I choose Kindroid or Nomi?

Choose the editing workflow you want to use: Kindroid provides a direct Response Directive and response examples; Nomi combines initial traits with editable context in Backstory+. Compare the same character premise, access tier and conversation task before paying.

Does Kindroid have an API, like Nomi?

Yes. Both have official API documentation. Kindroid’s group endpoints require a subscription. No live API access was tested by Atlas.

Which has better memory?

We have not benchmarked recall. The comparison establishes saved-context controls and group setup, rather than a memory-quality winner.

Can Nomi’s base traits be changed later?

Nomipedia says starting traits and interests cannot be edited directly after creation. Backstory+ can add or reinforce context, which is a different control.

Sources & scope.

Primary documents and recorded public offers underpin this article. We distinguish the source’s claims from our interpretation and suggested tests. No affiliate tracking or sponsorship is active.

Read the research methodology and editorial policy. Dates indicate a documentation review, not an account-level product test.