Thai-language AI character chat platform where users browse and create AI characters and role-play with them, and creators earn a token share based on how many messages others send.
A local Thai team (บริษัท ลองเล่น เอไอ จำกัด; "ลองเล่น" means "try playing") registered the domain in April 2025 and launched a beta in June, building a "Thai Character.AI". Users create their own AI characters and fantasy worlds, and the tokens spent when others chat with a character are shared with its creator, so creators have a reason to repost their characters to Facebook and other social platforms. On the strength of this UGC supply (the sitemap already has about 4,500 characters and 4,500 worlds) and Thailand's anime and romance role-play crowd, social makes up four-tenths of traffic. It monetizes with small token top-ups via PromptPay and TrueMoney (from ฿19), and this month it added an Android app (50k+ downloads).One-line growth story, compiled by us from public sources; speculation is marked separately below.
The main engine is creators bringing people in: anyone can build an AI character, and the more others chat with it the more tokens the creator earns, so character links get shared on social platforms like Facebook; direct visits are 43%, and second place is organic social.
The lubricant is daily and weekly leaderboards and small top-ups: the hot charts give new characters exposure, existing users come back directly to keep chatting, and token packs starting at ฿19 can be paid with PromptPay or TrueMoney.
By our estimate, nearly half of visits are existing users coming back to keep chatting; about 38 in every 100 visits click through to the checkout page.
Model: Lenny's racecar growth framework. Engine = creators bringing people in (anyone can build characters and worlds, tokens spent by others chatting are shared with the creator, and creators share characters on social platforms like Facebook to bring in new users); lubricant = daily and weekly leaderboards that give new characters exposure, and small top-ups from ฿19 using local Thai payments. The take-off event and the source of the first batch of users were not found.
Real data: 22,700 visits, -65.99% month over month, largest channel direct at 43.03%, top three channels ranked (direct, organic social, organic search), organic search 100% within search, 61 search terms in total, top search terms (longlenai, ลองเล่นเอไอ, longlen ai, lonaai, longlenal), referral site category (Technology 99.51%), countries (Thailand 98.82%, Vietnam 1.18%), bounce rate 21.66%, 5.61 pages per visit, average stay 3 minutes 10 seconds, outbound link destination pay.sn, all from the Similarweb public page, last month; the public page does not give the shares of the second and third channels, nor the names of the social platforms. 8,596 visits (about 8,600) to the Stripe checkout page come from the quietmoney checkout page data.
Supporting evidence: The Thai company LongLen AI Co., Ltd. registered the domain and launched the Beta in 2025; the sitemap has 4,509 character pages and 4,509 world pages; there is a creator earnings statistics page; token packs range from 300 tokens for ฿19 to 58,000 tokens for ฿1,699, with PromptPay, TrueMoney and credit card supported; in 2026 an Android app launched, with 50K+ downloads on Google Play; the official Facebook and Instagram accounts keep posting updates and calls to creators.
Inference: The two channel circles are real; how each route is split is an inference: direct 9,800 = existing users 7,500 + opening directly after seeing a social post 2,300; other channels 12,900 = seeing a character shared by a creator 8,500 + existing users coming back from social profiles or by searching the brand name 3,000 + finding it by searching characters 1,400. By ranking, organic social is larger than organic search, so the second circle gives more to social and less to search; the profile's social 42.7% and search 8.5% are from an earlier method and are not used here. That 5% of people who see a social post click through and 5% of searchers click through are guesses. Visits fell 66% month over month, so last month was a pullback period; the specific reason was not found.
Checkout: checkout.stripe.com
“Rankings/galleries: 'อันดับประจำวัน' (daily ranking), 'ดาวเด่น มาแรง' (trending), 'ตัวละครทั้งหมด' (all characters); creators get token share from messages othe”Quote from the site
New-user source: Social media (Organic social). Social 42.7% and direct 43.0% each make up about four-tenths, with the specific social platforms not listed; the site has daily/weekly rankings and popular characters, and the creator revenue share pushes authors to share their characters elsewhere. Search is only 8.5%, with brand keywords at 48.2% (longlenai, ลองเล่นเอไอ). Existing users return directly to keep chatting.
Top channel measured by Similarweb, others estimated by rank
| Attributable new users, last month | Share |
|---|---|
| SEO | 3% |
| 43% | |
| Other | 54% |
Direct + brand search 48%. Direct and brand search count as viral only when the site shows sharing mechanics; the largest share wins, under 5% means Other.
| Country | Share of visits |
|---|---|
| Thailand | 98.82% |
| Vietnam | 1.18% |
Mainly Thai, plus an English interface
61 keywords · brand ~48% · organic 100% / paid 0%
| Similarweb top 5 keywords |
|---|
| longlenai |
| ลองเล่นเอไอ |
| longlen ai |
| lonaai |
| longlenal |
Non-brand queries: ลองเล่นเอไอ, lonaai
| Category | Share of referrals |
|---|---|
| Computers Electronics and Technology - Other | 99.51% |
About 8,596 last month (Similarweb estimate), #425 of 606; the month before was 0. Opening checkout is not paying, so this is not revenue.
New-user source: Social media (Organic social). New users mostly come from YouTube, X, Reddit, TikTok and the like Social 42.7% and direct 43.0% each make up about four-tenths, with the specific social platforms not listed; the site has daily/weekly rankings and popular characters, and the creator revenue share pushes authors to share their characters elsewhere. Search is only 8.5%, with brand keywords at 48.2% (longlenai, ลองเล่นเอไอ). Existing users return directly to keep chatting.
Not found
Dossier confidence: medium. Parts 1 and 2 come from reading the site and Similarweb data; part 3 from public reporting, founder posts, Wayback snapshots, WHOIS and similar, with speculation marked.
The number: checkout visits is Similarweb's estimate of visits from the site to a checkout payment page; opening checkout is not paying, so this is not revenue. Small sites land on fixed buckets.
Data through last month. Page updated 2026-10-31.