Choosing between BaZi services
BaZi Alternatives: Three Axes to Compare Before You Choose
Most comparisons list features. This one lists what each service does on the parts that change a reading: whether the interpretation comes from software or from a named practitioner, whether your birthplace and true solar time are used, and which language and which system the reading follows. Every row below names the record, the section and the date behind it, and anything we could not verify is written as not verified.
How to read the tables
Four states are used, and the first word of every cell is the state. Yes: we verified that the service does this. Partly: verified with a qualification, or verified only as something the service says about itself. No: we verified that it does not. Not verified: we found no public evidence either way, so we claim nothing either way. A comparison that fills every cell usually guesses about half of them.
Axis 1: software or a named practitioner, and whether the service says where the line is
The useful question is not which is better. It is what is computed, what is interpreted, and whether the service says which is which. A chart is arithmetic — four pillars, ten gods, luck pillars, all determined by the birth data — and a reading is judgement layered on top of it. The table records whether each service states that split in its own words, because it is the checkable half of the software-versus-human question. One absence belongs in prose rather than in a row: none of the five AI-native services here publishes a named human practitioner behind its readings, and neither do we. That is an absence in our evidence rather than a verified fact about them, so it does not get a cell.
| Service | What we could verify |
|---|
| Mingsage (this site) | Partly. The engine-first split is published across our product pages — the free tool computes the chart in the browser, and the missing-hour page states that the engine drops the fourth pillar rather than filling it — but we have not gathered it into one methodology line the way two services below have.Recorded 2026-09-27 · source: competitive-improvement-plan-2026-09-27.md §WS-1 数据流 · competitive-improvement-plan-2026-09-27.md §WS-1 差异化 |
|---|
| openfate.ai | Yes. It publishes its own split in writing: a rules engine runs first, and the AI interprets without changing the deterministic inputs.Recorded 2026-09-27 · source: competitive-landscape-2026-09-27.md §3.3 第 6 条 |
|---|
| sajuapp.app | Yes. It publishes the argument that a chat model can write a reading but cannot calculate a chart, and its pricing page anchors the comparison at a human master's $55–58.Recorded 2026-09-27 · source: competitive-landscape-2026-09-27.md §3.3 第 6 条 · competitive-landscape-2026-09-27.md §4 D6 价格锚 |
|---|
| shen-shu.com | Not verified. We found no public statement from this service on where calculation stops and interpretation begins, so we claim nothing either way.Recorded 2026-09-27 · source: competitive-landscape-2026-09-27.md §6.1 未抓取的轴 |
|---|
| bloomprint.app | Not verified. We found no public statement from this service on where calculation stops and interpretation begins, so we claim nothing either way.Recorded 2026-09-27 · source: competitive-landscape-2026-09-27.md §6.1 未抓取的轴 |
|---|
| fatemaster.ai | Not verified. We found no public statement from this service on where calculation stops and interpretation begins, so we claim nothing either way.Recorded 2026-09-27 · source: competitive-landscape-2026-09-27.md §6.1 未抓取的轴 |
|---|
The two Yes rows quote the services' own sentences rather than paraphrasing them, and both state the split more plainly than we do. Our own row is Partly for that reason.
Axis 2: birthplace, true solar time, and the hour that was never recorded
True solar time is the gap between clock time and the sun's actual position at the birth place. It moves the two-hour boundaries that decide the hour pillar, and near midnight it can move the day pillar as well. A chart built without the birth city is not wrong about the date; it is unpinned about the hour. The offset is measured in minutes and the consequence is measured in pillars, because a boundary is a boundary.
| Service | What we could verify |
|---|
| Mingsage (this site) | Yes. The birth city is a required field rather than an optional extra, and the chart is computed with the true solar time correction for that place. The free tool prints the offset it applied, so the correction is visible rather than taken on trust.Recorded 2026-09-27 · source: free-chart.ts 校验层 place_required · competitive-improvement-plan-2026-09-27.md §WS-1 差异化 · free-chart-tool.tsx 真太阳时卡 |
|---|
| Chinese Fortune Calendar, bazi-calculator, Feng Shui Web (group) | Yes. Chinese Fortune Calendar, bazi-calculator and Feng Shui Web take a birthplace or a longitude and apply true solar time. They are the only three of the six long-running calculator sites where we could verify it.Recorded 2026-09-27 · source: competitive-landscape-2026-09-27.md §3.5 精度缺口 |
|---|
| Your Chinese Astrology, China Highlights, KarmaWeather (group) | No. Your Chinese Astrology asks for no birthplace and no longitude, China Highlights asks for a name and a date, and KarmaWeather asks for year, month and day. The date side of the chart still stands; the hour is simply not pinned to a place.Recorded 2026-09-27 · source: competitive-landscape-2026-09-27.md §3.5 精度缺口 · competitive-landscape-2026-09-27.md §3.5 精度缺口 |
|---|
| openfate.ai | Not verified. The public material we hold on this service covers its rectification page and its price; we found no public statement on birthplace input or true solar time, so we claim nothing either way.Recorded 2026-09-27 · source: competitive-landscape-2026-09-27.md §3.3 第 1 条 · competitive-landscape-2026-09-27.md §6.1 未抓取的轴 |
|---|
| sajuapp.app | Not verified. Our evidence records how it handles a missing birth hour; we found no public statement on birthplace input or true solar time, so we claim nothing either way.Recorded 2026-09-27 · source: competitive-landscape-2026-09-27.md §4 D8 · competitive-landscape-2026-09-27.md §6.1 未抓取的轴 |
|---|
| shen-shu.com | Not verified. Our evidence records its stated handling of a missing birth hour; we found no public statement on birthplace input or true solar time, so we claim nothing either way.Recorded 2026-09-27 · source: competitive-landscape-2026-09-27.md §4 D8 · competitive-landscape-2026-09-27.md §6.1 未抓取的轴 |
|---|
| bloomprint.app | Not verified. Our evidence records its stated handling of a missing birth hour; we found no public statement on birthplace input or true solar time, so we claim nothing either way.Recorded 2026-09-27 · source: competitive-landscape-2026-09-27.md §4 D8 · competitive-landscape-2026-09-27.md §6.1 未抓取的轴 |
|---|
| fatemaster.ai | Not verified. This service is marked not verified on birth-hour handling in our evidence; we found no public statement on birthplace input or true solar time, so we claim nothing either way.Recorded 2026-09-27 · source: competitive-landscape-2026-09-27.md §4 D8 · competitive-landscape-2026-09-27.md §6.1 未抓取的轴 |
|---|
2b. When the birth hour was never recorded: does the service say what it does?
This is the part of the comparison with the least room for interpretation, because the answer is a documented behaviour rather than a marketing line. Four different answers appear in this set. The one that costs accuracy is a default hour the customer is never told about; among the services we could verify, each one states its choice.
| Service | What we could verify |
|---|
| Mingsage (this site) | Yes. With no birth hour, the engine drops the hour pillar instead of filling one in, and everything the reading needed from the fourth pillar is marked open. It is a stated policy with a page of its own, not a silent fallback.Recorded 2026-09-27 · source: competitive-improvement-plan-2026-09-27.md §WS-1 差异化 · site-paths.ts PUBLIC_PATHS |
|---|
| openfate.ai | Yes. It publishes a dedicated birth time rectification page, priced at 75 points per run, and our recorded pass found it to be the only service in this set that offers one.Recorded 2026-09-27 · source: competitive-landscape-2026-09-27.md §3.3 第 1 条 · competitive-landscape-2026-09-27.md §4 D8 |
|---|
| sajuapp.app | Yes. It defaults to noon and tells the customer that it has done so.Recorded 2026-09-27 · source: competitive-landscape-2026-09-27.md §4 D8 |
|---|
| shen-shu.com | Yes. It fills 00:00 and states that precision is reduced.Recorded 2026-09-27 · source: competitive-landscape-2026-09-27.md §4 D8 |
|---|
| bloomprint.app | Yes. It declines orders that arrive without a birth time.Recorded 2026-09-27 · source: competitive-landscape-2026-09-27.md §4 D8 |
|---|
| fatemaster.ai | Not verified. We found no public statement on how this service handles a missing birth hour, so we claim nothing either way.Recorded 2026-09-27 · source: competitive-landscape-2026-09-27.md §4 D8 |
|---|
None of those four answers is wrong. A declared noon default and a declared reduction in precision keep a chart buildable and are honest about what changed; refusing the order is also honest, and costs a sale rather than an answer. The failure mode this table looks for is a default the customer is never told about, and among the services we could verify, none of them does that.
One service outside this table makes true solar time its own pitch and states that most BaZi apps leave it out. That is its own claim about the field rather than evidence we collected, so it is recorded here as a claim and not as a finding — which is the same reason the cells above say not verified instead of yes or no.
Axis 3: language coverage and which system the reading follows
Two different things get bundled into multi-language. Language coverage decides whether the reading arrives in a language you read natively. The system decides what the reading is about: Chinese four pillars and Korean saju share the four-pillar skeleton but differ in convention and vocabulary, and a Western astrology app is a different subject rather than a translation of this one. On coverage we claim no advantage, because there is none left to claim.
| Service | What we could verify |
|---|
| Mingsage (this site) | Yes. Two interface languages, English and Chinese, and nothing beyond that: more locales are planned rather than shipped. What we do hold that none of the five services in this table does is a classical Chinese source library, with the original texts online.Recorded 2026-09-27 · source: sitemap.ts LOCALES · competitive-landscape-2026-09-27.md §5 P2-4 · competitive-landscape-2026-09-27.md §4 D3 · competitive-landscape-2026-09-27.md §4 D3 |
|---|
| openfate.ai | Yes. It publishes five languages.Recorded 2026-09-27 · source: competitive-landscape-2026-09-27.md §3.3 第 4 条 |
|---|
| sajuapp.app | Yes. It publishes nine languages and localises its comparison pages into all nine. It is a Korean saju service, so saju vocabulary is its natural register.Recorded 2026-09-27 · source: competitive-landscape-2026-09-27.md §3.3 第 4 条 · competitive-landscape-2026-09-27.md §3.3 第 2 条 |
|---|
| shen-shu.com | Not verified. We found no public statement on this service's language coverage, so we claim nothing either way.Recorded 2026-09-27 · source: competitive-landscape-2026-09-27.md §6.1 未抓取的轴 |
|---|
| bloomprint.app | Not verified. We found no public statement on this service's language coverage, so we claim nothing either way.Recorded 2026-09-27 · source: competitive-landscape-2026-09-27.md §6.1 未抓取的轴 |
|---|
| fatemaster.ai | Not verified. We found no public statement on this service's language coverage, so we claim nothing either way.Recorded 2026-09-27 · source: competitive-landscape-2026-09-27.md §6.1 未抓取的轴 |
|---|
| bazi-calculator, KarmaWeather (group) | Yes. For scale: bazi-calculator publishes nineteen language versions with localised slugs, and KarmaWeather publishes six with translated paths. Both are far above the two languages we publish.Recorded 2026-09-27 · source: competitive-landscape-2026-09-27.md §3.3 第 4 条 / §4 D8 · competitive-landscape-2026-09-27.md §3.2 第 7 条 · competitive-landscape-2026-09-27.md §3.2 第 7 条 · competitive-landscape-2026-09-27.md §2.2 词簇地图 |
|---|
If language coverage decides your choice, this axis is already lost on our side, and the numbers above say which way to go. The system question is separate: saju and BaZi share the four-pillar skeleton, so a saju reading is not a mistranslation of a BaZi reading — it is the same structure handled by a different lineage.
The five services in this comparison, one by one
One card per service: what we verified, what the service says about itself, and what we could not establish. Nothing here is a recommendation and nothing here is a claim about quality.
openfate.ai
- What it is
- An AI-native BaZi service, one of the newer entrants that rank across our tracked query set.
- Published position we verified
- It publishes a rules-engine-first split — the engine computes, the AI interprets without changing the deterministic inputs — and runs a dedicated birth time rectification page at 75 points per run. It also open-sources a BaZi MCP server for developers.
- Price and refunds
- A published price of $26.99 per month, with a seven-day refund window.
- Missing birth hour
- Rectifies: it publishes a dedicated birth time rectification page instead of defaulting to an hour.
- Not verified
- Birthplace input and true solar time. Its pricing page is client-rendered, so a plain text fetch cannot read the price from it.
sajuapp.app
- What it is
- A Korean saju service, published in nine languages, that also writes publicly about building the product.
- Published position we verified
- It publishes the plainest version of the compute-first argument in this set: a chat model can write a reading but cannot calculate a chart. It also runs its own comparison pages, including a Korean saju app round-up and a saju-versus-Co-Star page — the same alternatives intent this page is written for.
- Price and refunds
- From $7.99, with a seven-day refund window and a pricing-page anchor of a human master's $55–58.
- Missing birth hour
- Defaults to noon and tells the customer that it has done so.
- Not verified
- Birthplace input and true solar time.
shen-shu.com
- What it is
- An AI-native service that ranks first for one of the reading-intent queries in our tracked set.
- Published position we verified
- It appears at the top of reading-intent results in our tracked query set; we found no public statement on where calculation stops and interpretation begins.
- Price and refunds
- A seven-day refund window. We did not verify its price.
- Missing birth hour
- Fills 00:00 and states that precision is reduced.
- Not verified
- Birthplace input, true solar time, language coverage, and where it draws the line between calculation and interpretation.
bloomprint.app
- What it is
- An AI-native service in the same 2024–2026 cohort as the rest of this group.
- Published position we verified
- It anchors its own comparison at consultations of €580 or more; we found no public statement on where calculation stops and interpretation begins.
- Price and refunds
- We did not verify its price. The €580 anchor above is what its own material states.
- Missing birth hour
- Declines orders that arrive without a birth time.
- Not verified
- Birthplace input, true solar time, language coverage and price.
fatemaster.ai
- What it is
- An AI-native service that appeared across seven of our fifteen tracked queries.
- Published position we verified
- We verified its search presence, and no published position on the axes above.
- Price and refunds
- Not verified. We found no public statement on its price or its refunds.
- Missing birth hour
- Not verified.
- Not verified
- The widest gap in this set: birthplace input, true solar time, language coverage, missing-hour handling, price and refunds.
Facts on this page are as of 27 September 2026 and come from one recorded review pass over what these services publish themselves. Each row above says what was verified and what was not. Nothing here is a claim about quality, and nothing here is an accusation.
Where we are not the best fit
A comparison page that only argues for one option is an advertisement. These are the cases where another service in this set is the better choice on the evidence above, followed by the cases where the three axes point to us.
Choose elsewhere if
- You want a named human practitioner standing behind the reading. We have none, and that gap does not close by writing more pages.
- You need the reading in Korean, Japanese, Portuguese or any language other than English or Chinese. sajuapp.app publishes nine languages and bazi-calculator publishes nineteen.
- You want the lowest price in this field. bazi-calculator publishes $3.80 per month, below every one of our tiers.
- You would rather not give a birthplace at all. Our engine asks for the city because the hour boundary depends on it, and the calculator sites that ask for less are easier for it.
- You want a service with a long track record. Several sites in this field have been online since the 1990s, and we do not have years behind us.
The three axes point to us if
- You want the birth city to be part of the calculation rather than an optional field, and you want to see the true solar time offset applied to your own data before you pay.
- You do not have a recorded birth hour and want it held open rather than filled in. Four of the five services in this comparison document their own handling, with different answers; ours is to drop the fourth pillar and mark the hour-dependent material as open.
- You want the interpretation layer tied to classical Chinese sources, with the original text kept alongside the reading rather than paraphrased away.
Each of the five services in this comparison does something we do not. This section exists so that the rest of the page is a comparison rather than a pitch.
BaZi alternatives: the questions that decide the choice
What is the difference between saju and bazi?
Both read four pillars built from the same year, month, day and hour positions. Saju is the Korean tradition and BaZi the Chinese one, so the skeleton is shared while the vocabulary, some conventions and the reference texts differ. A saju reading is not a mistranslation of a BaZi reading; it is the same structure handled by a different lineage, which is why a search for one often lands on the other.
Which BaZi service should I choose?
We will not answer that with a ranking, because the useful answer depends on which axis matters to you. If it is price, bazi-calculator publishes the lowest in this field at $3.80 per month. If it is language coverage, sajuapp.app publishes nine and bazi-calculator nineteen. If it is a named human practitioner, none of the AI-native services here publishes one as far as our evidence goes, and neither do we. If it is the missing birth hour, the table inside Axis 2 is the one to read.
Do any BaZi services use true solar time?
In the group we could verify, three of the six long-running calculator sites take a birthplace or a longitude and apply it, and three ask for no birthplace at all. For the newer AI-native services, including the five profiled here, we found no public statement either way, so this page records them as not verified rather than assuming.
What happens if I do not know my birth hour?
Four different documented answers appear in this set. sajuapp.app defaults to noon and says so; shen-shu.com fills 00:00 and says precision is reduced; bloomprint.app declines the order; openfate.ai publishes a dedicated rectification page at 75 points per run. Ours is to drop the hour pillar and mark the hour-dependent parts of the reading as open.
Is a software reading different from a reading by a human practitioner?
The difference runs in one direction when it comes to checking. A computed chart can be verified against your own birth data; an interpretation cannot be verified at all, whoever writes it. That is why this page compares the checkable half — birthplace handling, birth-hour handling, language coverage — and leaves the rest to your own reading of each service.
Why does this page say not verified instead of saying no?
Because a missing public statement is not evidence of a missing feature. Asserting that a service lacks something we never confirmed is how a comparison page ends up making claims it cannot support, and it is also how it ends up wrong. Every row here carries the date and the source we used, and where the source runs out the row says so.
Does Mingsage keep my birth details if I use the free tool?
The free chart runs in your browser: no account, no server-side chart record and no new storage table. The chart tool itself makes one request out: the city lookup, which goes through a geocoding proxy. We do not write that query to our application logs, and the place data it returns sits in an anonymous cache. Cloudflare's platform request logs are sampled at 1% and retained for up to 7 days. The chart itself is computed on your device.
How current is this comparison?
The facts here are as of 27 September 2026 and come from a single recorded review pass. Services change what they offer without announcing it, so check the service's own site before paying for anything.
Compare on the two things you can check yourself
Birthplace and birth hour are the two axes you can verify without trusting anyone: enter your city in the free tool and see the true solar time offset it applies, or leave the hour blank and see what a three-pillar chart does with it. If the reading is what you want next, it is built on that same computed chart.
Ordering takes about two minutes, no account needed; the report arrives as a PDF and stays available online.
For entertainment purposes only. This page compares published statements about other services, recorded on 27 September 2026; it is neither an endorsement nor a criticism of any of them, and we do not claim that our reading is more accurate than anyone else's. BaZi is reflective cultural guidance rather than a prediction of fixed fate, and it is not medical, legal or financial advice.