SYNTX AI is a subscription aggregator that puts third-party text, image, video and music models such as Seedance, Kling, Veo and Claude behind one Telegram bot and web dashboard, with generations paid for in internal tokens.
Used this tool? Rate it
Used this tool? Rate it
SYNTX AI (also written SYNTX.AI) is a subscription service that bundles more than 100 third-party AI models for text, images, video and music into a single Telegram bot and a web app. Instead of paying for each model separately, users of this all-in-one AI platform buy one plan, receive internal tokens, and spend those tokens on generations across the catalog.
The service is operated by SYNTX INTELLIGENCE – FZCO, a company registered in the United Arab Emirates on October 21, 2025, with its address in Dubai Silicon Oasis. The same company is named as the contractor in the public offer and as the party responsible for processing personal data in the privacy policy.
SYNTX AI does not present itself as a model developer. It is an access layer: the models in its catalog, such as Seedance, Kling, Veo, Nano Banana, Claude, Gemini and Suno, are developed by other companies, and the Seedance page, for example, credits ByteDance. The checked SYNTX pages do not describe any partnership, licensing or endorsement arrangement with those model developers; the site makes no statement on this point.
Three things shape most buying decisions:
The homepage model strip, as captured on September 27, 2026, names models including Nano Banana, Kling, Google Veo, GPT Image, Seedance 2.5, Seedream, Grok Imagine, Flux, Runway, Luma, Higgsfield Soul, HeyGen, Ideogram, Wan, Hailuo Minimax, Clarity, Elevenlabs, Magnific, Topaz AI, ChatGPT, Claude, Gemini, Grok, Deepseek, Perplexity, Qwen, Zhipu AI and Suno. The site describes the catalog as growing with "new models every week", so any list reflects only the day it was read.
Access is uneven across plans. The Basic card names Kling and Nano banana, and the homepage FAQ says the full functionality is available starting from the PRO plan. Elite and Ultra Elite are the tiers that add unlimited generation in selected models such as Midjourney, so a model appearing in the catalog does not mean every plan can use it.
LLM Studio groups the chat models. According to the homepage FAQ, all models in LLM Studio can access the internet and use current web information when answering. Basic includes basic access to text models, and higher plans widen the set of chat models that can be used without limits: Pro lists GPT 5 Nano, DeepSeek V3, Claude Haiku 4.5 and Gemini 3.5 Flash Lite, VIP adds Grok 4.3 and Qwen 3.7 Plus, Elite adds GPT 5.6 Luna and Gemini 3.5 Flash, and Ultra Elite adds Claude Sonnet 5.0 and Gemini 3.1 Pro.
Video generation is where the catalog is deepest. The Seedance 2.5 page is a representative example of how SYNTX packages a third-party model: it advertises clips of up to 30 seconds, up to 50 references, native lip-sync and 4K output, with a beta option to extend a clip to 180 seconds. The same page compares it with Seedance 2.0, which it lists at 15 seconds and 12 references, while the 2.5 references split into 30 image, 10 video and 10 audio.
Beyond generation, plan cards from Pro upward include creating GPT agents and enhancing video and image quality with Topaz. The knowledge base also lists face replacement, voice cloning, Stable Diffusion, Kling Try-On and the Clarity and Magnific upscalers among the tools on the platform.
The "Academy" section holds guides to the platform's tools and features, and the homepage points to a Knowledge Base, a YouTube channel and Telegram channels for further examples.
The official Seedance 2.5 page lays out the workflow in three steps:
On the web dashboard, the page says the video models sit under the Video tab: navigate to the Video tab and select a model from the dropdown menu on the right-hand side. Seedance 2.0 and 2.5 appear there alongside Kling, Veo and Grok Video.
Subscriptions paid through Stripe or Overpay renew automatically. Go to the "Profile" section and click "Cancel auto-renewal"; the subscription stays active until the end of the paid period, and auto-renewal can be switched back on later.
The official Seedance 2.5 page names four groups of users and the output each one gets. These are vendor-described examples rather than independent case studies:
SYNTX AI suits people who want several commercial models in one place and are comfortable working inside Telegram:
It fits less well for:
SYNTX AI runs in two places, and they are not equal:
The public offer also names @syntxaibot as the official and sole Syntx AI entry point and says other resources presenting themselves as Syntx AI are not recognized by the operator.
SYNTX AI pricing combines a subscription period with a token allowance. New users start on a FREE plan with 5 free tokens and 5 requests to language models, and the FREE plan has no time limit. Paid subscriptions run for 1, 3, 6 or 12 months.
The public price file behind the pricing section lists separate USD, EUR and RUB prices for each plan. The monthly entries below are the ones that include a token allowance, as captured on September 27, 2026:
| Plan | Monthly price (USD / EUR / RUB) | Tokens in that entry | Monthly allowance as described |
|---|---|---|---|
| Basic | $9.41 / €8.90 / 890 RUB | 260 | Up to 43 videos, 867 images or 52 audio tracks |
| Pro | $17.96 / €16.90 / 1690 RUB | 680 | Up to 113 videos, 2267 images or 136 audio tracks |
| VIP | $43.61 / €39.90 / 3990 RUB | 1700 | Up to 283 videos, 5667 images or 340 audio tracks |
| Elite | $65.46 / €59.90 / 5990 RUB | 2600 | Unlimited generation in Midjourney, Kling, Nano banana, Veo 3.1 and GPT Image; up to 520 songs |
| Ultra Elite | $125.40 / €119 / 11990 RUB | 3000 | Unlimited generation in Midjourney, Kling, Nano banana, Veo 3.1 and Sora; up to 600 songs |
For Basic, Pro, Elite and Ultra Elite, the same file also lists a cheaper monthly entry with 0 tokens (for example $8.46 for Basic); the checked pages do not explain what that variant includes. The Elite and Ultra Elite "unlimited" lines carry an asterisk that the plan data does not define, and the plan cards rate request volume from 3× Basic on Pro up to 15× Basic on Ultra Elite. The pricing toggle labels annual billing as "-15%".
A few rules change the real cost:
The refund policy is strict. A refund is possible only on the first purchase of a subscription and only if the service has not been used; any access after payment, including simply signing in, counts as use. Automatic renewal charges through Stripe and Overpay are not refundable regardless of use. Refund requests go to support on Telegram (@syntxhelp) and are reviewed within 21 business days. The offer also cites the 14-day withdrawal right under EU Directive 2011/83/EU, while stating that this right is lost once performance has begun with the customer's express consent.
The same Russian-language comparison of 17 aggregators that places SYNTX.AI among creator tools also lists services built around different needs:
SYNTX AI's own terms rule out automated access, so the API-first services in that list cover a need SYNTX does not, while Chad AI is positioned for everyday consumer chat.
SYNTX's knowledge base explains that filters are built-in restrictions that can make a model refuse a request and return no result. The platform mixes models with almost no built-in filtering and models with strict moderation, where breaking the rules can lead to restrictions or blocked access. The knowledge base groups Ideogram v3.0 and Sora among the strictest models, and lists face replacement, voice cloning, Stable Diffusion and Topaz AI among the minimal-filter tools. Even in that minimal-filter group, it lists illegal material, any form of child exploitation, and NSFW and sensitive content as prohibited, and it places responsibility for results entirely on the user.
The public offer says all exclusive rights to materials the customer creates with Syntx AI belong to the customer, and that the customer may use them for personal or commercial purposes and transfer them to third parties. It also requires that created materials comply with UAE law and do not infringe third-party intellectual property.
Yes. New accounts get 5 free tokens and 5 requests to language models, and the FREE plan has no time limit. The company describes this as enough to try the platform and understand how it works.
No. The Telegram bot carries the full feature set, the web version does not yet include everything, and an account created with Google can only be used on the web.
No. Tokens stay on the account after a subscription ends, but they can be spent only while a subscription is active, and they are added to new tokens on renewal.
Generally not. The offer allows a refund only on a first subscription purchase that has not been used, and any sign-in after payment counts as use. Automatic renewal charges through Stripe and Overpay are not refundable.
The public offer says the customer holds the rights to materials created with Syntx AI and may use them for personal or commercial purposes, provided they comply with UAE law and do not infringe third-party rights.
