Hypit
Officially listedAn open-source tool where coding agents clone viral video workflows at scale.
Hypit
Hypit is an open-source AI video cloning tool from Shenzhen-based Kaxi Yongliu Technology that gives coding agents like Claude Code and Codex a language for describing video (SVML) plus a compile-and-run runtime, letting them batch-replicate winning short-video formats into hundreds of variants in one command.
Core Capabilities
- Video cloning: drop in a reference video and the agent reverse-engineers the complete workflow, not just a one-off render
- Anchored to words, not seconds: subtitles/B-roll/effects bind to individual words in the script, so changing lines, languages, or presenters re-lays-out automatically — a genuine mechanism innovation
- Batch variants: one command produces 10/50/100+ hook variants, replicating viral structures at scale
- SVML video DSL: five layers (script/assets/subtitles/voiceover/compositing), described as "HTML for video"
- Model freedom + BYOK: compatible with Claude Code/Codex, bring your own model keys, free core code with no watermark
Use Cases Built for developers and growth marketing teams using Claude Code/Codex who have validated a short-video structure and need to mass-produce multi-language/multi-placement variants; also for technical teams researching "agent-driven video pipelines." Pure creators should look elsewhere.
Unique Advantages Compared with closed-source talking-head/ad tools like Arcads and Higgsfield, Hypit clones the "workflow" rather than a one-off video, is open and controllable, and lets you swap models. But official examples are only 18–26 seconds, and long-video capability is unproven.
Editorial Review A real open-source project, not a shell: 11,204 GitHub stars, 357 paid subscriptions, and Stripe-verified MRR of $5,734. But the marketing runs ahead of the evidence: the brand wall lists Google/Meta and 14 companies total with no corroboration, while primary interviews show initial customers were just two small AI startups, Cluely and Stella; the "100M views" figure has no source; the license is NOASSERTION (modified Apache 2.0 with commercial restrictions), inconsistent with the "Open Source" claim; and costs cannot be estimated upfront — Issue #270 reports a user burning 10 million tokens in the analysis phase with zero videos produced. Best for developer/agent users; start with cheaper models and set budget caps.
Pricing
### 💰 定价模式：免费增值（开源核心免费 + Credits 订阅） **起步价**：免费（开源核心免费 + BYOK 自带 Key） #### 主要方案 - **开源核心 + BYOK**：$0 - 代码免费、无座位费、无 per-render 抽成、无水印，模型服务需自付 - **Starter**：$9.90/月 - 每月 1,800 Credits - **Pro**：$39.90/月（原价 $49.90）- 每月 10,000 Credits，官方推荐 - **Ultra**：$99.90/月（原价 $129.90）- 每月 27,000 Credits - **Custom / Business**：联系 official@hypit.ai - 多租户商用授权与企业定制 #### 试用/其他信息 Credits 充值包按量付费（2,000 credits $15 起，充值额度永不过期）；定价页真实路径为 /commercial/pricing/（官网 /pricing 为 404）。 — Visit website
FAQ
Is Hypit free?
The open-source core and BYOK (bring your own key) are free to use, with no watermark and no per-render cut. The HypiHub hosted subscription runs Starter $9.90 / Pro $39.90 / Ultra $99.90 per month.
What can Hypit be used for?
It is an AI video cloning tool for coding agents like Claude Code and Codex, able to reverse-engineer the complete workflow of a viral video and batch-produce 10/50/100+ hook variants.
Can Hypit clone any viral video?
You can drop in a reference video and the agent reverse-engineers the workflow, but official examples are only 18–26 seconds; minute-long video capability is unproven, and costs need your own verification (they cannot be calculated upfront).
Does using Hypit require programming skills?
Yes. Hypit is not a no-code tool; workflows are written in SVML source and managed by a coding agent. Teams without developers or an agent in the loop should use a hosted platform instead.
How is Hypit different from Arcads and Higgsfield?
The core difference is that Hypit clones the "workflow" rather than a one-off video, and is open-source, controllable, and BYOK model-swappable; Arcads and Higgsfield are closed-source SaaS priced at $39–$129+/month.
Is Hypit a true open-source project?
The code is public with 11,204 stars on GitHub, but the license is NOASSERTION (modified Apache 2.0 plus commercial restrictions, banning multi-tenant SaaS and resale), so it is not OSI-standard open source.
Which company makes Hypit?
Shenzhen Kaxi Yongliu Technology Co., Ltd. (unified social credit code 91440300MAK83W9E6H), registered on 2026-02-04 in Nanshan, Shenzhen, with legal representative Chen Yudong and an angel round on 2026-04-02; disclosed proactively in the site's ToS/privacy policy, showing good transparency.
Can Hypit's costs be estimated upfront?
No. The official Credits documentation admits per-model multipliers are "deliberately not published"; GitHub Issue #270 reports a user burning 10 million tokens during the analysis phase with zero videos, so set a budget cap first.
Can we trust Hypit's claim that Google, Meta, and other big companies use it?
Doubtful. The brand wall has no corroboration, and 357 subscriptions can hardly support enterprise-scale procurement at tech giants; primary LinkedIn interviews show initial customers were just two small AI startups, Cluely and Stella — more likely individual employees.
What is Hypit's user reputation like?
Chinese social media is full of near-identical promotional posts, while English channels (Product Hunt/Hacker News/Reddit/G2) show almost no genuine user discussion; but subscriptions and revenue are real, with Stripe-verified MRR of $5,734 and $9,555 cumulative.