aiga_

aiga_ is a browser AI RPG platform from AI WORX LTD that began life as “AI Game Agent”. You either play a world somebody else built or write your own from a short prompt, and then play it in one of four shapes: Interactive Story (a guided, chapter-like narrative where choices shape what happens next), Open World / RPG (freeform exploration with characters, locations, factions, objectives, inventory and evolving world state), Single Objective (one focused scenario built around one clear goal), and Kingpin (a competitive mode a compatible world can run, where you build power and reputation and are ranked against other runs). It plays solo, in a private group with friends, or as a community game anybody can join — and it reaches out into Discord, Telegram and X, where a server or a following can vote on the next move.

It is the rare entry that is held back by its build rather than by its design.

What It Scores

aiga_ scores 2.9 out of 5, eleventh of the twenty-five platforms on our benchmark, level with MacerAI.

Signature Design — 4 is the high mark, and Kingpin is most of it. Kingpin is a mode, not a place — a compatible world can run it, and the worlds that do get a leaderboard. Play in one and your job is to build as much reputation as you can and not die, against other people doing the same thing; the ranking is both within that world and sitewide. Nothing else in this directory turns an AI RPG into a competition you can lose to a stranger. Leaderboards attach to the worlds running the mode rather than collapsing into one global number: each run reports its authoritative progression score, the world keeps the best and current ranked entries, and a seasonal view can reset the competitive window without deleting the underlying game. Ordinary story and RPG worlds still show popularity and play counts, but those are discovery stats, not a competitive board — a distinction most platforms would have blurred. Around it sit an unusual number of finished ideas: image continuity that holds your hero’s face across scenes and across twelve art styles from a photo or a sketch you upload, sixteen interface languages, and a “Living Game Book” that keeps each world’s own history.

Mechanical Depth — 3, Memory & Continuity — 3, Player Agency — 3 and Longevity — 3 are four straight neutral scores, and they are neutral for four different reasons.

The systems are there and they are well thought out — objectives, inventory, factions, locations, evolving world state — and they are genuinely good when they work as they are supposed to. That last clause is the review. Memory is regular for a world that is a prompt plus an engine: the Living Game Book holds relationships and alliances, and it neither surprised us nor let us down. Agency is mid-table for the ordinary reasons — you are mostly free to do what you want, but the game still tries to send you off to the quest it would prefer you were doing, and there are one or two remnants of puppeting left in it, though nothing like the worst of the field. Longevity gets real help from Kingpin, which is the one thing here designed to still matter at hour twenty, and loses it again to the bugs.

Determinism & Fairness — 2 and NPC Fidelity — 2 are where it actually loses the composite, and neither is the usual complaint.

Fairness would be a 3 — the neutral score, outcomes neither earned nor arbitrary — except that the bugs reach the ledger. Rewards and purchases can come out wrong. That is a different failure from the one this axis was built to catch: the field’s normal fairness problem is a narrator that agrees with everything you say, and aiga_’s is a game that agreed, adjudicated, and then miscredited you. The player experience is the same either way — what you did stopped reliably determining what you got — which is why it scores where it scores.

NPC Fidelity is the weakest axis here. Characters do not have a very distinct voice, and a bug sometimes has one character speak another character’s lines. Casts flattening into one voice is the most common failure on this axis across the whole field; this is that failure with a defect underneath it rather than a model limitation, which is at least the more fixable version.

The Shape This Score Makes

Every other entry on this board scores what it scores because of what somebody decided to build. aiga_ is the first that scores what it scores because of what somebody has not finished debugging — and the two are worth telling apart, because they do not predict the same future.

The design is above its score. Four ambitious game modes, a competitive layer with a real ranking, cross-platform voting that turns a Discord server into a table, image continuity, sixteen languages: that is a lot of product, and the parts of it we could exercise cleanly were good. What is underneath it is a build where the rewards do not always land and the characters occasionally trade dialogue. Those are not architecture. They are the kind of thing that gets closed by a quarter of patches — which is exactly what we expect here, and exactly why this entry will be worth re-scoring rather than filing.

There is one honest caveat about the competition itself. In community games everyone can vote, and each turn counts the first N votes it receives — where N is set by your plan, from 5 on the free tier to 50 on Gold. That is a defensible way to price server load, and it is also a paid tier buying more weight in a shared story, which is worth knowing before you organise a community around it.

Modes and Where You Play Them

Mode	What it is
Interactive Story	A guided, chapter-like narrative where choices shape what happens next
Open World / RPG	Freeform exploration with characters, locations, factions, objectives, inventory and evolving world state
Single Objective	A focused scenario built around completing one clear goal
Kingpin	A competitive mode a compatible world can run: players build power and reputation and are ranked against other runs

Solo play is private and self-paced. Private groups invite friends and family with configurable voting on shared choices. Community games are open worlds anybody can join and vote in. Beyond the website, aiga_ runs games from Discord servers, takes timed story polls on X, and manages games from Telegram chats with voting handled on the web.

There is no store app on either platform. The site ships a web app manifest — standalone display, maskable icons at 192 and 512 pixels — so it installs to a home screen and opens without browser chrome, which is the same route ten other platforms here take.

Pricing

Free to start, with subscriptions and one-off credit top-ups. The unusual part is what credits are for: text play is included with every paid plan — creating worlds, starting games, taking story turns and sending chat messages spend no credits at all — and credits are spent on generated images, video clips and other media. The free tier is metered in turns instead: five text-only turns a day, no card required. Verified on aiga.io, 8 September 2026:

Plan	Price	Media credits/month	Roughly	Votes counted per turn
Free	$0	0	Text only, 5 turns/day	5
Explorer	$8.99/mo	0	Text play, no media	10
Bronze	$11.99/mo	3,000	~30 images or ~7 video clips	10
Silver	$23.99/mo	14,000	~140 images or ~35 video clips	20
Gold	$59.99/mo	50,000	~500 images or ~125 video clips	50

Additional credits can be bought at any time. Plans can be changed at any time — upgrades are prorated against the remaining billing cycle, downgrades apply from the next one — and cancelling leaves access running to the end of the paid period. Refunds are case by case through support. An enterprise tier exists for larger community-voting allowances and is quoted on contact.

Explorer is the tier worth understanding: $8.99 buys unmetered text play and nothing else. If you never generate an image, it is the whole product.

No underlying model or provider is named anywhere on the site.

Verdict

aiga_ has more ideas than most of this board and less polish than nearly all of it. Kingpin is a genuinely new answer to the question of why you would come back on day thirty, the mode design is thought through, and the platform reach — Discord, Telegram, X, sixteen languages — is real product rather than a landing-page list. What it has not done is make the bugs stop touching the parts that count: the rewards, the purchases, and the characters’ own lines. There is a lot of potential here, and it is still a bit unpolished. We expect improvements next quarter and intend to re-score it. 2.9 out of 5.

Product names and trademarks are the property of their respective owners. Arcanum is not affiliated with aiga_.